{
"cells": [
{
"cell_type": "markdown",
"id": "intro",
"metadata": {},
"source": [
"# ECG HR/HRV Pretraining — Fixed Version\n",
"\n",
"Key fixes applied:\n",
"1. **Sliding-window extraction** — turns 22 records into thousands of training samples\n",
"2. **Correct `key_dim`** — set to `d_model // num_heads` (16) instead of 64\n",
"3. **Pre-LN Transformer** — more stable training\n",
"4. **Dropout** added for regularization on small dataset\n",
"5. **Cosine LR schedule** with warmup\n",
"6. **Proper evaluation** with train/val split"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "cell_01_imports",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1775488166.145818 216784 port.cc:153] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.\n",
"I0000 00:00:1775488166.228778 216784 cpu_feature_guard.cc:227] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
"To enable the following instructions: AVX2 AVX_VNNI AVX_VNNI_INT8 AVX_NE_CONVERT FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1775488167.495536 216784 port.cc:153] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"TF version: 2.21.0\n"
]
}
],
"source": [
"import os\n",
"import datetime\n",
"import numpy as np\n",
"import tensorflow as tf\n",
"import wfdb\n",
"import joblib\n",
"\n",
"print(\"TF version:\", tf.__version__)"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "cell_02_download",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Using cache directory: ../../db/afdb\n",
"True\n",
"Found 25 records in AFDB.\n",
"\n",
"=== Loading full record 00735 ===\n",
"Skipping 00735: sampto must be greater than sampfrom\n",
"\n",
"=== Loading full record 03665 ===\n",
"Skipping 03665: sampto must be greater than sampfrom\n",
"\n",
"=== Loading full record 04015 ===\n",
"Loaded 9205760 samples (10.23 hours)\n",
"\n",
"=== Loading full record 04043 ===\n",
"Loaded 9205760 samples (10.23 hours)\n",
"\n",
"=== Loading full record 04048 ===\n",
"Loaded 9205760 samples (10.23 hours)\n",
"\n",
"=== Loading full record 04126 ===\n",
"Loaded 9205760 samples (10.23 hours)\n",
"\n",
"=== Loading full record 04746 ===\n",
"Loaded 9205760 samples (10.23 hours)\n",
"\n",
"=== Loading full record 04908 ===\n",
"Loaded 9205760 samples (10.23 hours)\n",
"\n",
"=== Loading full record 04936 ===\n",
"Loaded 9205760 samples (10.23 hours)\n",
"\n",
"=== Loading full record 05091 ===\n",
"Loaded 9205760 samples (10.23 hours)\n",
"\n",
"=== Loading full record 05121 ===\n",
"Loaded 9205760 samples (10.23 hours)\n",
"\n",
"=== Loading full record 05261 ===\n",
"Loaded 9205760 samples (10.23 hours)\n",
"\n",
"=== Loading full record 06426 ===\n",
"Loaded 9205760 samples (10.23 hours)\n",
"\n",
"=== Loading full record 06453 ===\n",
"Loaded 8325000 samples (9.25 hours)\n",
"\n",
"=== Loading full record 06995 ===\n",
"Loaded 9205760 samples (10.23 hours)\n",
"\n",
"=== Loading full record 07162 ===\n",
"Loaded 9205760 samples (10.23 hours)\n",
"\n",
"=== Loading full record 07859 ===\n",
"Loaded 9205760 samples (10.23 hours)\n",
"\n",
"=== Loading full record 07879 ===\n",
"Loaded 9205760 samples (10.23 hours)\n",
"\n",
"=== Loading full record 07910 ===\n",
"Loaded 9205760 samples (10.23 hours)\n",
"\n",
"=== Loading full record 08215 ===\n",
"Loaded 9205760 samples (10.23 hours)\n",
"\n",
"=== Loading full record 08219 ===\n",
"Loaded 9205760 samples (10.23 hours)\n",
"\n",
"=== Loading full record 08378 ===\n",
"Loaded 9205760 samples (10.23 hours)\n",
"\n",
"=== Loading full record 08405 ===\n",
"Loaded 9205760 samples (10.23 hours)\n",
"\n",
"=== Loading full record 08434 ===\n",
"Loaded 9205760 samples (10.23 hours)\n",
"\n",
"=== Loading full record 08455 ===\n",
"Loaded 9205760 samples (10.23 hours)\n",
"\n",
"Downloaded 23 full-length ECG signals\n"
]
},
{
"data": {
"text/plain": [
"['../../db/afdb_signals.pkl']"
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"import wfdb\n",
"import os\n",
"\n",
"def download_afdb_records(max_records=30, cache_dir=\"../../db/afdb\"):\n",
" \"\"\"\n",
" Downloads and loads full 10-hour AFDB ECG signals.\n",
" \n",
" - Downloads the entire AFDB database ONCE into cache_dir\n",
" - Loads full-length signals (≈10 hours, ~9.2M samples)\n",
" - Skips broken records gracefully\n",
" \"\"\"\n",
" os.makedirs(cache_dir, exist_ok=True)\n",
" print(f\"Using cache directory: {cache_dir}\")\n",
" print(os.path.exists(cache_dir))\n",
" db_name = \"afdb\"\n",
"\n",
" records = wfdb.get_record_list(db_name)\n",
" print(f\"Found {len(records)} records in AFDB.\")\n",
" records = records[:max_records]\n",
"\n",
" # Download database once if missing\n",
" if not any(fname.endswith(\".hea\") for fname in os.listdir(cache_dir)):\n",
" print(\"Downloading AFDB database (one-time)...\")\n",
" wfdb.dl_database(db_name, dl_dir=cache_dir)\n",
" print(\"Download complete. Using local cache.\")\n",
"\n",
" signals = []\n",
"\n",
" for rec in records:\n",
" try:\n",
" print(f\"\\n=== Loading full record {rec} ===\")\n",
" record = wfdb.rdrecord(cache_dir + \"/\" + rec)\n",
"\n",
" sig = record.p_signal[:, 0].astype(\"float32\")\n",
" print(f\"Loaded {sig.shape[0]} samples ({sig.shape[0] / 250 / 3600:.2f} hours)\")\n",
"\n",
" signals.append(sig)\n",
"\n",
" except Exception as e:\n",
" print(f\"Skipping {rec}: {e}\")\n",
" continue\n",
"\n",
" return signals\n",
"\n",
"\n",
"# Example usage:\n",
"raw_signals = download_afdb_records(max_records=30)\n",
"print(\"\\nDownloaded\", len(raw_signals), \"full-length ECG signals\")\n",
"\n",
"import joblib\n",
"joblib.dump(raw_signals, \"../../db/afdb_signals.pkl\")\n"
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "cell_04_extract",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Loaded raw signals: 23\n",
"Total windows: 177 (seq_len=128)\n",
"HR shape: (177, 128) HRV shape: (177, 128)\n",
"HR range: [105.6, 195.5] mean=150.7\n",
"HRV range: [0.0033, 0.2253] mean=0.1023\n"
]
}
],
"source": [
"import numpy as np\n",
"import joblib\n",
"from joblib import Parallel, delayed\n",
"from scipy.signal import butter, filtfilt, find_peaks\n",
"import tensorflow as tf\n",
"\n",
"# ============================================================\n",
"# 1. FAST R-PEAK DETECTOR (Pan-Tompkins style, vectorized)\n",
"# ============================================================\n",
"\n",
"def bandpass_filter(ecg, fs=250, low=5, high=15):\n",
" b, a = butter(1, [low/(fs/2), high/(fs/2)], btype='band')\n",
" return filtfilt(b, a, ecg)\n",
"\n",
"def fast_rpeak_detector(ecg, fs=250):\n",
" x = bandpass_filter(ecg, fs)\n",
" x2 = x * x\n",
" win = int(0.15 * fs)\n",
" mwa = np.convolve(x2, np.ones(win)/win, mode='same')\n",
" peaks, _ = find_peaks(mwa, distance=int(0.25 * fs))\n",
" return peaks\n",
"\n",
"# ============================================================\n",
"# 2. HR + HRV (RMSSD) EXTRACTION with configurable window/step\n",
"# ============================================================\n",
"\n",
"def fast_hr_hrv_from_rpeaks(rpeaks, fs=250, window_sec=60, step_sec=60):\n",
" \"\"\"\n",
" Extract HR and HRV (RMSSD) from R-peaks using a sliding window.\n",
" step_sec < window_sec creates overlapping windows = more samples.\n",
" \"\"\"\n",
" if len(rpeaks) < 3:\n",
" return None, None\n",
"\n",
" rr_intervals = np.diff(rpeaks) / fs\n",
" rr_intervals = rr_intervals[(rr_intervals > 0.3) & (rr_intervals < 2.0)]\n",
"\n",
" if len(rr_intervals) < 3:\n",
" return None, None\n",
"\n",
" rr_times = rpeaks[1:len(rr_intervals)+1] / fs\n",
" t_max = rr_times[-1]\n",
" times = np.arange(0, t_max - window_sec + 1, step_sec)\n",
"\n",
" hr = np.full(len(times), np.nan, dtype=\"float32\")\n",
" hrv = np.full(len(times), np.nan, dtype=\"float32\")\n",
"\n",
" for i, t0 in enumerate(times):\n",
" t1 = t0 + window_sec\n",
" mask = (rr_times >= t0) & (rr_times < t1)\n",
" if np.sum(mask) < 3:\n",
" continue\n",
" rr_win = rr_intervals[mask]\n",
" hr_val = 60.0 / np.mean(rr_win)\n",
" if 30 <= hr_val <= 220:\n",
" hr[i] = hr_val\n",
" diff_rr = np.diff(rr_win)\n",
" if len(diff_rr) > 0:\n",
" hrv[i] = np.sqrt(np.mean(diff_rr**2))\n",
"\n",
" valid = ~np.isnan(hr)\n",
" if np.sum(valid) < 10:\n",
" return None, None\n",
"\n",
" return hr[valid], hrv[valid]\n",
"\n",
"def extract_hr_hrv_fast(ecg, fs=250):\n",
" rpeaks = fast_rpeak_detector(ecg, fs)\n",
" return fast_hr_hrv_from_rpeaks(rpeaks, fs)\n",
"\n",
"# ============================================================\n",
"# 3. SLIDING WINDOW: extract many fixed-length subsequences\n",
"# from each record's HR/HRV time series.\n",
"# This is the KEY fix — turns 22 records into many samples.\n",
"# ============================================================\n",
"\n",
"T = 128 # sequence length (minutes) — shorter = more samples\n",
"STEP = 32 # step size between windows — smaller = more overlap = more samples\n",
"\n",
"def extract_windows(hr_seq, hrv_seq, T=128, step=32):\n",
" \"\"\"Slice a HR/HRV time series into fixed-length windows.\"\"\"\n",
" windows_hr, windows_hrv = [], []\n",
" n = len(hr_seq)\n",
" for start in range(0, n - T + 1, step):\n",
" windows_hr.append(hr_seq[start:start+T])\n",
" windows_hrv.append(hrv_seq[start:start+T])\n",
" return windows_hr, windows_hrv\n",
"\n",
"# ============================================================\n",
"# 4. LOAD & PROCESS\n",
"# ============================================================\n",
"\n",
"raw_signals = joblib.load(\"../../db/afdb_signals.pkl\")\n",
"print(\"Loaded raw signals:\", len(raw_signals))\n",
"\n",
"# Extract HR/HRV per record (in parallel)\n",
"results = Parallel(n_jobs=-1)(\n",
" delayed(extract_hr_hrv_fast)(sig) for sig in raw_signals\n",
")\n",
"\n",
"# Slice into windows\n",
"hr_list, hrv_list = [], []\n",
"for hr, hrv in results:\n",
" if hr is None:\n",
" continue\n",
" wins_hr, wins_hrv = extract_windows(hr, hrv, T=T, step=STEP)\n",
" hr_list.extend(wins_hr)\n",
" hrv_list.extend(wins_hrv)\n",
"\n",
"hr_arr = np.array(hr_list, dtype=\"float32\") # (N, T)\n",
"hrv_arr = np.array(hrv_list, dtype=\"float32\") # (N, T)\n",
"\n",
"# Activity is still zeros (placeholder for future wearable data)\n",
"activity_arr = np.zeros((hr_arr.shape[0], hr_arr.shape[1], 3), dtype=\"float32\")\n",
"\n",
"print(f\"Total windows: {hr_arr.shape[0]} (seq_len={T})\")\n",
"\n",
"# ============================================================\n",
"# 5. IMPUTE NaNs\n",
"# ============================================================\n",
"\n",
"def impute_nan(arr):\n",
" m = np.nanmean(arr, axis=1, keepdims=True)\n",
" return np.where(np.isnan(arr), m, arr)\n",
"\n",
"hr_arr = impute_nan(hr_arr)\n",
"hrv_arr = impute_nan(hrv_arr)\n",
"\n",
"print(\"HR shape:\", hr_arr.shape, \" HRV shape:\", hrv_arr.shape)\n",
"print(f\"HR range: [{hr_arr.min():.1f}, {hr_arr.max():.1f}] mean={hr_arr.mean():.1f}\")\n",
"print(f\"HRV range: [{hrv_arr.min():.4f}, {hrv_arr.max():.4f}] mean={hrv_arr.mean():.4f}\")"
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "cell_05_dataset",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Train: 150 | Val: 27\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"W0000 00:00:1775488180.836681 216784 gpu_device.cc:2459] TensorFlow was not built with CUDA kernel binaries compatible with compute capability 12.0a. CUDA kernels will be jit-compiled from PTX, which could take 30 minutes or longer.\n",
"W0000 00:00:1775488180.842899 216784 gpu_device.cc:2459] TensorFlow was not built with CUDA kernel binaries compatible with compute capability 12.0a. CUDA kernels will be jit-compiled from PTX, which could take 30 minutes or longer.\n",
"I0000 00:00:1775488181.072186 216784 gpu_device.cc:2043] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 9226 MB memory: -> device: 0, name: NVIDIA GeForce RTX 5070 Ti Laptop GPU, pci bus id: 0000:01:00.0, compute capability: 12.0a\n"
]
}
],
"source": [
"# ============================================================\n",
"# 6. TRAIN / VAL SPLIT\n",
"# ============================================================\n",
"\n",
"N = hr_arr.shape[0]\n",
"idx = np.random.permutation(N)\n",
"split = int(0.85 * N)\n",
"train_idx, val_idx = idx[:split], idx[split:]\n",
"\n",
"hr_train = hr_arr[train_idx]\n",
"hr_val = hr_arr[val_idx]\n",
"\n",
"print(f\"Train: {hr_train.shape[0]} | Val: {hr_val.shape[0]}\")\n",
"\n",
"# ============================================================\n",
"# 7. TF DATASET WITH MASKING\n",
"# ============================================================\n",
"BATCH_SIZE = 32\n",
"MASK_RATIO = 0.15\n",
"SEQ_LEN = T # 128\n",
"\n",
"def make_ds(hr, shuffle=True):\n",
" ds = tf.data.Dataset.from_tensor_slices(hr) # each element: (SEQ_LEN,)\n",
" if shuffle:\n",
" ds = ds.shuffle(2048)\n",
" ds = ds.batch(BATCH_SIZE).prefetch(tf.data.AUTOTUNE)\n",
" return ds\n",
"\n",
"def add_mask(hr_batch, mask_ratio=MASK_RATIO):\n",
" # hr_batch: (B, SEQ_LEN)\n",
" mask = tf.cast(tf.random.uniform(tf.shape(hr_batch)) < mask_ratio, tf.float32)\n",
" hr_masked = hr_batch * (1.0 - mask)\n",
" return {\n",
" \"hr\": hr_masked, # input to model\n",
" \"target\": hr_batch, # full target\n",
" \"mask\": mask, # for evaluation only\n",
" }\n",
"\n",
"def map_to_model_inputs(batch):\n",
" inputs = {\"hr\": batch[\"hr\"]} # (B, SEQ_LEN)\n",
" targets = batch[\"target\"] # (B, SEQ_LEN)\n",
" return inputs, targets # NO sample_weight here\n",
"\n",
"train_ds = (\n",
" make_ds(hr_train)\n",
" .map(add_mask, num_parallel_calls=tf.data.AUTOTUNE)\n",
" .map(map_to_model_inputs, num_parallel_calls=tf.data.AUTOTUNE)\n",
")\n",
"\n",
"val_ds = (\n",
" make_ds(hr_val, shuffle=False)\n",
" .map(add_mask, num_parallel_calls=tf.data.AUTOTUNE)\n",
" .map(map_to_model_inputs, num_parallel_calls=tf.data.AUTOTUNE)\n",
")\n",
"\n",
"\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "cell_06_model",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"Model: \"hr_encoder\"\n",
"\n"
],
"text/plain": [
"\u001b[1mModel: \"hr_encoder\"\u001b[0m\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┓\n",
"┃ Layer (type) ┃ Output Shape ┃ Param # ┃ Connected to ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━┩\n",
"│ hr (InputLayer) │ (None, 128) │ 0 │ - │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ reshape (Reshape) │ (None, 128, 1) │ 0 │ hr[0][0] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ hr_norm │ (None, 128, 1) │ 3 │ reshape[0][0] │\n",
"│ (Normalization) │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dense (Dense) │ (None, 128, 64) │ 128 │ hr_norm[0][0] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ add (Add) │ (None, 128, 64) │ 0 │ dense[0][0] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ layer_normalization │ (None, 128, 64) │ 128 │ add[0][0] │\n",
"│ (LayerNormalizatio… │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ multi_head_attenti… │ (None, 128, 64) │ 16,640 │ layer_normalizat… │\n",
"│ (MultiHeadAttentio… │ │ │ layer_normalizat… │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dropout_1 (Dropout) │ (None, 128, 64) │ 0 │ multi_head_atten… │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ add_1 (Add) │ (None, 128, 64) │ 0 │ add[0][0], │\n",
"│ │ │ │ dropout_1[0][0] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ layer_normalizatio… │ (None, 128, 64) │ 128 │ add_1[0][0] │\n",
"│ (LayerNormalizatio… │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dense_1 (Dense) │ (None, 128, 256) │ 16,640 │ layer_normalizat… │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dropout_2 (Dropout) │ (None, 128, 256) │ 0 │ dense_1[0][0] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dense_2 (Dense) │ (None, 128, 64) │ 16,448 │ dropout_2[0][0] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dropout_3 (Dropout) │ (None, 128, 64) │ 0 │ dense_2[0][0] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ add_2 (Add) │ (None, 128, 64) │ 0 │ add_1[0][0], │\n",
"│ │ │ │ dropout_3[0][0] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ layer_normalizatio… │ (None, 128, 64) │ 128 │ add_2[0][0] │\n",
"│ (LayerNormalizatio… │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ multi_head_attenti… │ (None, 128, 64) │ 16,640 │ layer_normalizat… │\n",
"│ (MultiHeadAttentio… │ │ │ layer_normalizat… │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dropout_5 (Dropout) │ (None, 128, 64) │ 0 │ multi_head_atten… │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ add_3 (Add) │ (None, 128, 64) │ 0 │ add_2[0][0], │\n",
"│ │ │ │ dropout_5[0][0] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ layer_normalizatio… │ (None, 128, 64) │ 128 │ add_3[0][0] │\n",
"│ (LayerNormalizatio… │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dense_3 (Dense) │ (None, 128, 256) │ 16,640 │ layer_normalizat… │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dropout_6 (Dropout) │ (None, 128, 256) │ 0 │ dense_3[0][0] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dense_4 (Dense) │ (None, 128, 64) │ 16,448 │ dropout_6[0][0] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dropout_7 (Dropout) │ (None, 128, 64) │ 0 │ dense_4[0][0] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ add_4 (Add) │ (None, 128, 64) │ 0 │ add_3[0][0], │\n",
"│ │ │ │ dropout_7[0][0] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ layer_normalizatio… │ (None, 128, 64) │ 128 │ add_4[0][0] │\n",
"│ (LayerNormalizatio… │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ multi_head_attenti… │ (None, 128, 64) │ 16,640 │ layer_normalizat… │\n",
"│ (MultiHeadAttentio… │ │ │ layer_normalizat… │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dropout_9 (Dropout) │ (None, 128, 64) │ 0 │ multi_head_atten… │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ add_5 (Add) │ (None, 128, 64) │ 0 │ add_4[0][0], │\n",
"│ │ │ │ dropout_9[0][0] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ layer_normalizatio… │ (None, 128, 64) │ 128 │ add_5[0][0] │\n",
"│ (LayerNormalizatio… │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dense_5 (Dense) │ (None, 128, 256) │ 16,640 │ layer_normalizat… │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dropout_10 │ (None, 128, 256) │ 0 │ dense_5[0][0] │\n",
"│ (Dropout) │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dense_6 (Dense) │ (None, 128, 64) │ 16,448 │ dropout_10[0][0] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dropout_11 │ (None, 128, 64) │ 0 │ dense_6[0][0] │\n",
"│ (Dropout) │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ add_6 (Add) │ (None, 128, 64) │ 0 │ add_5[0][0], │\n",
"│ │ │ │ dropout_11[0][0] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ layer_normalizatio… │ (None, 128, 64) │ 128 │ add_6[0][0] │\n",
"│ (LayerNormalizatio… │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dense_7 (Dense) │ (None, 128, 1) │ 65 │ layer_normalizat… │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ reshape_1 (Reshape) │ (None, 128) │ 0 │ dense_7[0][0] │\n",
"└─────────────────────┴───────────────────┴────────────┴───────────────────┘\n",
"\n"
],
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1mLayer (type) \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m Param #\u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mConnected to \u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━┩\n",
"│ hr (\u001b[38;5;33mInputLayer\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ - │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ reshape (\u001b[38;5;33mReshape\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m1\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ hr[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ hr_norm │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m1\u001b[0m) │ \u001b[38;5;34m3\u001b[0m │ reshape[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"│ (\u001b[38;5;33mNormalization\u001b[0m) │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dense (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m128\u001b[0m │ hr_norm[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ add (\u001b[38;5;33mAdd\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ dense[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ layer_normalization │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m128\u001b[0m │ add[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"│ (\u001b[38;5;33mLayerNormalizatio…\u001b[0m │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ multi_head_attenti… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m16,640\u001b[0m │ layer_normalizat… │\n",
"│ (\u001b[38;5;33mMultiHeadAttentio…\u001b[0m │ │ │ layer_normalizat… │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dropout_1 (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ multi_head_atten… │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ add_1 (\u001b[38;5;33mAdd\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ add[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m], │\n",
"│ │ │ │ dropout_1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ layer_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m128\u001b[0m │ add_1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"│ (\u001b[38;5;33mLayerNormalizatio…\u001b[0m │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dense_1 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m256\u001b[0m) │ \u001b[38;5;34m16,640\u001b[0m │ layer_normalizat… │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dropout_2 (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m256\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ dense_1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dense_2 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m16,448\u001b[0m │ dropout_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dropout_3 (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ dense_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ add_2 (\u001b[38;5;33mAdd\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ add_1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m], │\n",
"│ │ │ │ dropout_3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ layer_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m128\u001b[0m │ add_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"│ (\u001b[38;5;33mLayerNormalizatio…\u001b[0m │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ multi_head_attenti… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m16,640\u001b[0m │ layer_normalizat… │\n",
"│ (\u001b[38;5;33mMultiHeadAttentio…\u001b[0m │ │ │ layer_normalizat… │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dropout_5 (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ multi_head_atten… │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ add_3 (\u001b[38;5;33mAdd\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ add_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m], │\n",
"│ │ │ │ dropout_5[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ layer_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m128\u001b[0m │ add_3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"│ (\u001b[38;5;33mLayerNormalizatio…\u001b[0m │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dense_3 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m256\u001b[0m) │ \u001b[38;5;34m16,640\u001b[0m │ layer_normalizat… │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dropout_6 (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m256\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ dense_3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dense_4 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m16,448\u001b[0m │ dropout_6[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dropout_7 (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ dense_4[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ add_4 (\u001b[38;5;33mAdd\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ add_3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m], │\n",
"│ │ │ │ dropout_7[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ layer_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m128\u001b[0m │ add_4[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"│ (\u001b[38;5;33mLayerNormalizatio…\u001b[0m │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ multi_head_attenti… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m16,640\u001b[0m │ layer_normalizat… │\n",
"│ (\u001b[38;5;33mMultiHeadAttentio…\u001b[0m │ │ │ layer_normalizat… │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dropout_9 (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ multi_head_atten… │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ add_5 (\u001b[38;5;33mAdd\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ add_4[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m], │\n",
"│ │ │ │ dropout_9[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ layer_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m128\u001b[0m │ add_5[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"│ (\u001b[38;5;33mLayerNormalizatio…\u001b[0m │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dense_5 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m256\u001b[0m) │ \u001b[38;5;34m16,640\u001b[0m │ layer_normalizat… │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dropout_10 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m256\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ dense_5[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"│ (\u001b[38;5;33mDropout\u001b[0m) │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dense_6 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m16,448\u001b[0m │ dropout_10[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dropout_11 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ dense_6[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"│ (\u001b[38;5;33mDropout\u001b[0m) │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ add_6 (\u001b[38;5;33mAdd\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ add_5[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m], │\n",
"│ │ │ │ dropout_11[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ layer_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m128\u001b[0m │ add_6[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"│ (\u001b[38;5;33mLayerNormalizatio…\u001b[0m │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dense_7 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m1\u001b[0m) │ \u001b[38;5;34m65\u001b[0m │ layer_normalizat… │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ reshape_1 (\u001b[38;5;33mReshape\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ dense_7[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"└─────────────────────┴───────────────────┴────────────┴───────────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
" Total params: 150,276 (587.02 KB)\n",
"\n"
],
"text/plain": [
"\u001b[1m Total params: \u001b[0m\u001b[38;5;34m150,276\u001b[0m (587.02 KB)\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
" Trainable params: 150,273 (587.00 KB)\n",
"\n"
],
"text/plain": [
"\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m150,273\u001b[0m (587.00 KB)\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
" Non-trainable params: 3 (16.00 B)\n",
"\n"
],
"text/plain": [
"\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m3\u001b[0m (16.00 B)\n"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# ============================================================\n",
"# 8. MODEL — Pre-LN Transformer with fixed key_dim\n",
"#\n",
"# Fixes vs original:\n",
"# * key_dim = d_model // num_heads (was wrongly set to d_model)\n",
"# * Pre-LN (LayerNorm BEFORE attention/FFN) — more stable\n",
"# * Dropout added for regularisation\n",
"# ============================================================\n",
"\n",
"def transformer_block_pre_ln(x, d_model, num_heads, dropout_rate=0.1):\n",
" \"\"\"Pre-LN Transformer block: LN → Attn → Add → LN → FFN → Add\"\"\"\n",
" key_dim = d_model // num_heads # FIX: was d_model (64) — now 16\n",
"\n",
" # Self-attention sub-layer\n",
" x_ln = tf.keras.layers.LayerNormalization()(x)\n",
" attn = tf.keras.layers.MultiHeadAttention(\n",
" num_heads=num_heads,\n",
" key_dim=key_dim,\n",
" dropout=dropout_rate,\n",
" )(x_ln, x_ln)\n",
" attn = tf.keras.layers.Dropout(dropout_rate)(attn)\n",
" x = tf.keras.layers.Add()([x, attn])\n",
"\n",
" # Feed-forward sub-layer\n",
" x_ln = tf.keras.layers.LayerNormalization()(x)\n",
" ff = tf.keras.layers.Dense(d_model * 4, activation=\"gelu\")(x_ln)\n",
" ff = tf.keras.layers.Dropout(dropout_rate)(ff)\n",
" ff = tf.keras.layers.Dense(d_model)(ff)\n",
" ff = tf.keras.layers.Dropout(dropout_rate)(ff)\n",
" x = tf.keras.layers.Add()([x, ff])\n",
"\n",
" return x\n",
"\n",
"\n",
"def build_encoder(seq_len=128, d_model=64, num_heads=4, num_layers=3, dropout_rate=0.1):\n",
"\n",
" inputs = {\n",
" \"hr\": tf.keras.Input(shape=(seq_len,), name=\"hr\"),\n",
" }\n",
"\n",
" hr = tf.keras.layers.Reshape((seq_len, 1))(inputs[\"hr\"])\n",
" hr = tf.keras.layers.Normalization(axis=-1, name=\"hr_norm\")(hr)\n",
" x = tf.keras.layers.Dense(d_model)(hr)\n",
"\n",
" pos = tf.range(start=0, limit=seq_len, delta=1)\n",
" pos_emb = tf.keras.layers.Embedding(input_dim=seq_len, output_dim=d_model)(pos)\n",
" x = x + pos_emb\n",
"\n",
" for _ in range(num_layers):\n",
" x = transformer_block_pre_ln(x, d_model, num_heads, dropout_rate)\n",
"\n",
" x = tf.keras.layers.LayerNormalization()(x)\n",
" x = tf.keras.layers.Dense(1)(x)\n",
" x = tf.keras.layers.Reshape((seq_len,))(x) # (B, SEQ_LEN)\n",
"\n",
" return tf.keras.Model(inputs=inputs, outputs=x, name=\"hr_encoder\")\n",
"\n",
"SEQ_LEN = T\n",
"encoder = build_encoder(seq_len=SEQ_LEN)\n",
"\n",
"encoder.get_layer(\"hr_norm\").adapt(hr_train.reshape(-1, SEQ_LEN, 1))\n",
"hr_input = tf.keras.Input(shape=(SEQ_LEN,), name=\"hr\")\n",
"hr_out = encoder({\"hr\": hr_input}) # encoder expects dict\n",
"pretrain_model = tf.keras.Model(inputs={\"hr\": hr_input}, outputs=hr_out)\n",
"\n",
"encoder.summary()\n"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "cell_07_train",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 1/2000\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"I0000 00:00:1775488202.298999 217006 service.cc:153] XLA service 0x736d18051c50 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
"I0000 00:00:1775488202.299025 217006 service.cc:161] StreamExecutor [0]: NVIDIA GeForce RTX 5070 Ti Laptop GPU, Compute Capability 12.0a (Driver: 13.1.0; Runtime: 12.9.0; Toolkit: 12.5.0; DNN: 9.15.1)\n",
"I0000 00:00:1775488202.514739 217006 dump_mlir_util.cc:269] disabling MLIR crash reproducer, set env var `MLIR_CRASH_REPRODUCER_DIRECTORY` to enable.\n",
"I0000 00:00:1775488203.499197 217006 cuda_dnn.cc:461] Loaded cuDNN version 91501\n",
"I0000 00:00:1775488203.801267 217006 dot_merger.cc:481] Merging Dots in computation: a_inference_one_step_on_data_12425__.129\n",
"I0000 00:00:1775488206.166946 217213 subprocess_compilation.cc:348] ptxas warning : Registers are spilled to local memory in function 'gemm_fusion_dot_88', 4 bytes spill stores, 4 bytes spill loads\n",
"\n",
"I0000 00:00:1775488207.561793 217217 subprocess_compilation.cc:348] ptxas warning : Registers are spilled to local memory in function 'gemm_fusion_MatMul_1_94', 8 bytes spill stores, 8 bytes spill loads\n",
"\n",
"I0000 00:00:1775488207.619158 217221 subprocess_compilation.cc:348] ptxas warning : Registers are spilled to local memory in function 'gemm_fusion_MatMul_1_94', 8 bytes spill stores, 8 bytes spill loads\n",
"\n",
"I0000 00:00:1775488208.512668 217222 subprocess_compilation.cc:348] ptxas warning : Registers are spilled to local memory in function 'gemm_fusion_dot_90', 8 bytes spill stores, 8 bytes spill loads\n",
"\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m1/5\u001b[0m \u001b[32m━━━━\u001b[0m\u001b[37m━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m1:28\u001b[0m 22s/step - loss: 150.3906 - mae: 150.8906 - mape: 100.3618 - r2: -179.7240 - rmse: 151.3317"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"I0000 00:00:1775488218.440856 217006 device_compiler.h:208] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m3/5\u001b[0m \u001b[32m━━━━━━━━━━━━\u001b[0m\u001b[37m━━━━━━━━\u001b[0m \u001b[1m0s\u001b[0m 50ms/step - loss: 150.5223 - mae: 151.0223 - mape: 100.3836 - r2: -200.2092 - rmse: 151.4214 "
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"I0000 00:00:1775488219.635567 217008 dot_merger.cc:481] Merging Dots in computation: a_inference_one_step_on_data_12425__.129\n",
"I0000 00:00:1775488220.829347 217699 subprocess_compilation.cc:348] ptxas warning : Registers are spilled to local memory in function 'gemm_fusion_MatMul_1_94', 8 bytes spill stores, 8 bytes spill loads\n",
"\n",
"I0000 00:00:1775488220.877257 217707 subprocess_compilation.cc:348] ptxas warning : Registers are spilled to local memory in function 'gemm_fusion_MatMul_1_94', 8 bytes spill stores, 8 bytes spill loads\n",
"\n",
"I0000 00:00:1775488221.182671 217709 subprocess_compilation.cc:348] ptxas warning : Registers are spilled to local memory in function 'gemm_fusion_dot_90', 8 bytes spill stores, 8 bytes spill loads\n",
"\n",
"I0000 00:00:1775488223.166414 217698 subprocess_compilation.cc:348] ptxas warning : Registers are spilled to local memory in function 'gemm_fusion_dot_88', 4 bytes spill stores, 4 bytes spill loads\n",
"\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4s/step - loss: 150.6222 - mae: 151.1222 - mape: 100.3693 - r2: -203.6965 - rmse: 151.5147 "
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"I0000 00:00:1775488235.434749 217006 dot_merger.cc:481] Merging Dots in computation: a_inference_one_step_on_data_13576__.32\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m45s\u001b[0m 6s/step - loss: 150.7719 - mae: 151.2719 - mape: 100.3336 - r2: -203.4171 - rmse: 151.6644 - val_loss: 150.2096 - val_mae: 150.7096 - val_mape: 100.0370 - val_r2: -239.8611 - val_rmse: 151.0527\n",
"Epoch 2/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 61ms/step - loss: 149.8401 - mae: 150.3401 - mape: 99.7046 - r2: -203.9198 - rmse: 150.7480 - val_loss: 148.0532 - val_mae: 148.5532 - val_mape: 98.5837 - val_r2: -233.1358 - val_rmse: 148.9289\n",
"Epoch 3/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 60ms/step - loss: 147.8672 - mae: 148.3672 - mape: 98.3746 - r2: -197.4692 - rmse: 148.8085 - val_loss: 145.4578 - val_mae: 145.9578 - val_mape: 96.8473 - val_r2: -225.1193 - val_rmse: 146.3571\n",
"Epoch 4/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 145.5655 - mae: 146.0655 - mape: 96.8357 - r2: -191.3476 - rmse: 146.5247 - val_loss: 143.1236 - val_mae: 143.6236 - val_mape: 95.2950 - val_r2: -217.9738 - val_rmse: 144.0261\n",
"Epoch 5/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 143.4181 - mae: 143.9181 - mape: 95.4081 - r2: -185.6998 - rmse: 144.3798 - val_loss: 141.3708 - val_mae: 141.8708 - val_mape: 94.1396 - val_r2: -212.6230 - val_rmse: 142.2555\n",
"Epoch 6/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 141.7901 - mae: 142.2901 - mape: 94.3347 - r2: -180.6790 - rmse: 142.7346 - val_loss: 140.2757 - val_mae: 140.7757 - val_mape: 93.4175 - val_r2: -209.3074 - val_rmse: 141.1472\n",
"Epoch 7/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 140.7533 - mae: 141.2533 - mape: 93.6527 - r2: -178.2268 - rmse: 141.6835 - val_loss: 139.5555 - val_mae: 140.0555 - val_mape: 92.9459 - val_r2: -207.1208 - val_rmse: 140.4115\n",
"Epoch 8/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 140.0603 - mae: 140.5603 - mape: 93.1978 - r2: -179.0238 - rmse: 140.9784 - val_loss: 139.1296 - val_mae: 139.6296 - val_mape: 92.6654 - val_r2: -205.8401 - val_rmse: 139.9789\n",
"Epoch 9/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 139.6080 - mae: 140.1080 - mape: 92.8989 - r2: -176.3879 - rmse: 140.5213 - val_loss: 138.8104 - val_mae: 139.3104 - val_mape: 92.4558 - val_r2: -204.8794 - val_rmse: 139.6534\n",
"Epoch 10/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 139.2447 - mae: 139.7447 - mape: 92.6584 - r2: -173.8785 - rmse: 140.1551 - val_loss: 138.5219 - val_mae: 139.0219 - val_mape: 92.2631 - val_r2: -204.0305 - val_rmse: 139.3652\n",
"Epoch 11/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 138.9675 - mae: 139.4675 - mape: 92.4750 - r2: -175.2208 - rmse: 139.8758 - val_loss: 138.3109 - val_mae: 138.8109 - val_mape: 92.1231 - val_r2: -203.4066 - val_rmse: 139.1530\n",
"Epoch 12/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 138.7300 - mae: 139.2300 - mape: 92.3168 - r2: -173.3985 - rmse: 139.6381 - val_loss: 138.0902 - val_mae: 138.5902 - val_mape: 91.9754 - val_r2: -202.7616 - val_rmse: 138.9333\n",
"Epoch 13/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 138.5132 - mae: 139.0132 - mape: 92.1727 - r2: -172.4429 - rmse: 139.4211 - val_loss: 137.9244 - val_mae: 138.4244 - val_mape: 91.8655 - val_r2: -202.2731 - val_rmse: 138.7666\n",
"Epoch 14/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 138.3178 - mae: 138.8178 - mape: 92.0424 - r2: -173.8298 - rmse: 139.2260 - val_loss: 137.7452 - val_mae: 138.2452 - val_mape: 91.7459 - val_r2: -201.7503 - val_rmse: 138.5881\n",
"Epoch 15/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 138.1325 - mae: 138.6325 - mape: 91.9196 - r2: -171.8903 - rmse: 139.0401 - val_loss: 137.5712 - val_mae: 138.0712 - val_mape: 91.6300 - val_r2: -201.2420 - val_rmse: 138.4143\n",
"Epoch 16/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 137.9553 - mae: 138.4553 - mape: 91.8012 - r2: -171.9976 - rmse: 138.8637 - val_loss: 137.4100 - val_mae: 137.9100 - val_mape: 91.5227 - val_r2: -200.7715 - val_rmse: 138.2531\n",
"Epoch 17/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 137.7845 - mae: 138.2845 - mape: 91.6876 - r2: -170.7602 - rmse: 138.6931 - val_loss: 137.2457 - val_mae: 137.7457 - val_mape: 91.4133 - val_r2: -200.2926 - val_rmse: 138.0890\n",
"Epoch 18/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 137.6182 - mae: 138.1182 - mape: 91.5767 - r2: -169.4431 - rmse: 138.5274 - val_loss: 137.0829 - val_mae: 137.5829 - val_mape: 91.3047 - val_r2: -199.8199 - val_rmse: 137.9267\n",
"Epoch 19/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 137.4536 - mae: 137.9536 - mape: 91.4671 - r2: -169.4535 - rmse: 138.3630 - val_loss: 136.9302 - val_mae: 137.4302 - val_mape: 91.2030 - val_r2: -199.3758 - val_rmse: 137.7741\n",
"Epoch 20/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 137.2928 - mae: 137.7928 - mape: 91.3598 - r2: -169.4703 - rmse: 138.2027 - val_loss: 136.7715 - val_mae: 137.2715 - val_mape: 91.0969 - val_r2: -198.9171 - val_rmse: 137.6163\n",
"Epoch 21/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 137.1317 - mae: 137.6317 - mape: 91.2525 - r2: -169.9029 - rmse: 138.0420 - val_loss: 136.6142 - val_mae: 137.1142 - val_mape: 90.9922 - val_r2: -198.4605 - val_rmse: 137.4591\n",
"Epoch 22/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 136.9713 - mae: 137.4713 - mape: 91.1458 - r2: -170.0603 - rmse: 137.8817 - val_loss: 136.4606 - val_mae: 136.9606 - val_mape: 90.8898 - val_r2: -198.0163 - val_rmse: 137.3060\n",
"Epoch 23/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 136.8138 - mae: 137.3138 - mape: 91.0408 - r2: -168.2158 - rmse: 137.7247 - val_loss: 136.3066 - val_mae: 136.8066 - val_mape: 90.7872 - val_r2: -197.5711 - val_rmse: 137.1523\n",
"Epoch 24/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 136.6516 - mae: 137.1516 - mape: 90.9326 - r2: -168.6576 - rmse: 137.5631 - val_loss: 136.1478 - val_mae: 136.6478 - val_mape: 90.6812 - val_r2: -197.1127 - val_rmse: 136.9939\n",
"Epoch 25/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 136.4960 - mae: 136.9960 - mape: 90.8290 - r2: -167.4549 - rmse: 137.4077 - val_loss: 135.9926 - val_mae: 136.4926 - val_mape: 90.5778 - val_r2: -196.6654 - val_rmse: 136.8392\n",
"Epoch 26/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 136.3360 - mae: 136.8360 - mape: 90.7224 - r2: -167.0987 - rmse: 137.2481 - val_loss: 135.8393 - val_mae: 136.3393 - val_mape: 90.4756 - val_r2: -196.2238 - val_rmse: 136.6862\n",
"Epoch 27/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 136.1778 - mae: 136.6778 - mape: 90.6171 - r2: -167.1766 - rmse: 137.0901 - val_loss: 135.6804 - val_mae: 136.1804 - val_mape: 90.3696 - val_r2: -195.7664 - val_rmse: 136.5276\n",
"Epoch 28/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 136.0182 - mae: 136.5182 - mape: 90.5105 - r2: -168.2049 - rmse: 136.9314 - val_loss: 135.5235 - val_mae: 136.0235 - val_mape: 90.2651 - val_r2: -195.3159 - val_rmse: 136.3713\n",
"Epoch 29/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 135.8579 - mae: 136.3579 - mape: 90.4038 - r2: -169.7614 - rmse: 136.7714 - val_loss: 135.3669 - val_mae: 135.8669 - val_mape: 90.1607 - val_r2: -194.8661 - val_rmse: 136.2149\n",
"Epoch 30/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 135.6987 - mae: 136.1987 - mape: 90.2977 - r2: -165.8456 - rmse: 136.6126 - val_loss: 135.2096 - val_mae: 135.7096 - val_mape: 90.0558 - val_r2: -194.4153 - val_rmse: 136.0581\n",
"Epoch 31/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 135.5369 - mae: 136.0369 - mape: 90.1897 - r2: -166.5257 - rmse: 136.4514 - val_loss: 135.0502 - val_mae: 135.5502 - val_mape: 89.9495 - val_r2: -193.9589 - val_rmse: 135.8991\n",
"Epoch 32/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 135.3763 - mae: 135.8763 - mape: 90.0827 - r2: -164.8562 - rmse: 136.2911 - val_loss: 134.8907 - val_mae: 135.3907 - val_mape: 89.8431 - val_r2: -193.5028 - val_rmse: 135.7401\n",
"Epoch 33/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 135.2155 - mae: 135.7155 - mape: 89.9757 - r2: -168.6120 - rmse: 136.1306 - val_loss: 134.7294 - val_mae: 135.2294 - val_mape: 89.7357 - val_r2: -193.0420 - val_rmse: 135.5792\n",
"Epoch 34/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 135.0529 - mae: 135.5529 - mape: 89.8671 - r2: -163.6752 - rmse: 135.9689 - val_loss: 134.5685 - val_mae: 135.0685 - val_mape: 89.6284 - val_r2: -192.5830 - val_rmse: 135.4187\n",
"Epoch 35/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 134.8901 - mae: 135.3901 - mape: 89.7587 - r2: -163.1738 - rmse: 135.8064 - val_loss: 134.4065 - val_mae: 134.9065 - val_mape: 89.5204 - val_r2: -192.1209 - val_rmse: 135.2570\n",
"Epoch 36/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 134.7261 - mae: 135.2261 - mape: 89.6495 - r2: -164.6276 - rmse: 135.6426 - val_loss: 134.2442 - val_mae: 134.7442 - val_mape: 89.4122 - val_r2: -191.6591 - val_rmse: 135.0952\n",
"Epoch 37/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 134.5636 - mae: 135.0636 - mape: 89.5410 - r2: -163.9350 - rmse: 135.4811 - val_loss: 134.0808 - val_mae: 134.5808 - val_mape: 89.3033 - val_r2: -191.1947 - val_rmse: 134.9323\n",
"Epoch 38/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 134.3969 - mae: 134.8969 - mape: 89.4301 - r2: -162.4398 - rmse: 135.3145 - val_loss: 133.9152 - val_mae: 134.4152 - val_mape: 89.1929 - val_r2: -190.7243 - val_rmse: 134.7671\n",
"Epoch 39/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 134.2316 - mae: 134.7316 - mape: 89.3198 - r2: -161.8407 - rmse: 135.1498 - val_loss: 133.7496 - val_mae: 134.2496 - val_mape: 89.0825 - val_r2: -190.2548 - val_rmse: 134.6019\n",
"Epoch 40/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 48ms/step - loss: 134.0639 - mae: 134.5639 - mape: 89.2080 - r2: -164.0134 - rmse: 134.9825 - val_loss: 133.5833 - val_mae: 134.0833 - val_mape: 88.9716 - val_r2: -189.7834 - val_rmse: 134.4360\n",
"Epoch 41/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 133.8954 - mae: 134.3954 - mape: 89.0956 - r2: -164.4084 - rmse: 134.8148 - val_loss: 133.4151 - val_mae: 133.9151 - val_mape: 88.8595 - val_r2: -189.3079 - val_rmse: 134.2683\n",
"Epoch 42/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 133.7263 - mae: 134.2263 - mape: 88.9830 - r2: -162.0217 - rmse: 134.6460 - val_loss: 133.2475 - val_mae: 133.7475 - val_mape: 88.7477 - val_r2: -188.8343 - val_rmse: 134.1011\n",
"Epoch 43/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 133.5573 - mae: 134.0573 - mape: 88.8703 - r2: -160.0262 - rmse: 134.4775 - val_loss: 133.0782 - val_mae: 133.5782 - val_mape: 88.6349 - val_r2: -188.3564 - val_rmse: 133.9322\n",
"Epoch 44/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 133.3883 - mae: 133.8883 - mape: 88.7575 - r2: -163.3167 - rmse: 134.3093 - val_loss: 132.9076 - val_mae: 133.4076 - val_mape: 88.5212 - val_r2: -187.8756 - val_rmse: 133.7621\n",
"Epoch 45/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 133.2154 - mae: 133.7154 - mape: 88.6425 - r2: -159.4305 - rmse: 134.1367 - val_loss: 132.7355 - val_mae: 133.2355 - val_mape: 88.4065 - val_r2: -187.3911 - val_rmse: 133.5904\n",
"Epoch 46/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 49ms/step - loss: 133.0423 - mae: 133.5423 - mape: 88.5272 - r2: -159.4157 - rmse: 133.9641 - val_loss: 132.5629 - val_mae: 133.0629 - val_mape: 88.2913 - val_r2: -186.9061 - val_rmse: 133.4183\n",
"Epoch 47/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 132.8685 - mae: 133.3685 - mape: 88.4112 - r2: -160.0604 - rmse: 133.7908 - val_loss: 132.3888 - val_mae: 132.8888 - val_mape: 88.1753 - val_r2: -186.4173 - val_rmse: 133.2447\n",
"Epoch 48/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 132.6939 - mae: 133.1939 - mape: 88.2947 - r2: -158.9992 - rmse: 133.6171 - val_loss: 132.2145 - val_mae: 132.7145 - val_mape: 88.0592 - val_r2: -185.9287 - val_rmse: 133.0709\n",
"Epoch 49/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 132.5201 - mae: 133.0201 - mape: 88.1789 - r2: -158.9890 - rmse: 133.4437 - val_loss: 132.0383 - val_mae: 132.5383 - val_mape: 87.9416 - val_r2: -185.4352 - val_rmse: 132.8951\n",
"Epoch 50/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 48ms/step - loss: 132.3418 - mae: 132.8418 - mape: 88.0601 - r2: -157.8184 - rmse: 133.2659 - val_loss: 131.8619 - val_mae: 132.3619 - val_mape: 87.8241 - val_r2: -184.9418 - val_rmse: 132.7192\n",
"Epoch 51/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 132.1652 - mae: 132.6652 - mape: 87.9425 - r2: -160.1343 - rmse: 133.0895 - val_loss: 131.6838 - val_mae: 132.1838 - val_mape: 87.7053 - val_r2: -184.4447 - val_rmse: 132.5416\n",
"Epoch 52/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 131.9864 - mae: 132.4864 - mape: 87.8234 - r2: -157.6082 - rmse: 132.9114 - val_loss: 131.5043 - val_mae: 132.0043 - val_mape: 87.5857 - val_r2: -183.9440 - val_rmse: 132.3626\n",
"Epoch 53/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 52ms/step - loss: 131.8075 - mae: 132.3075 - mape: 87.7039 - r2: -156.5170 - rmse: 132.7332 - val_loss: 131.3243 - val_mae: 131.8243 - val_mape: 87.4657 - val_r2: -183.4426 - val_rmse: 132.1830\n",
"Epoch 54/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 131.6267 - mae: 132.1267 - mape: 87.5836 - r2: -155.3791 - rmse: 132.5528 - val_loss: 131.1434 - val_mae: 131.6434 - val_mape: 87.3451 - val_r2: -182.9395 - val_rmse: 132.0027\n",
"Epoch 55/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 131.4449 - mae: 131.9449 - mape: 87.4622 - r2: -158.3859 - rmse: 132.3721 - val_loss: 130.9609 - val_mae: 131.4609 - val_mape: 87.2234 - val_r2: -182.4327 - val_rmse: 131.8207\n",
"Epoch 56/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 131.2621 - mae: 131.7621 - mape: 87.3406 - r2: -154.9383 - rmse: 132.1893 - val_loss: 130.7777 - val_mae: 131.2777 - val_mape: 87.1013 - val_r2: -181.9245 - val_rmse: 131.6379\n",
"Epoch 57/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 131.0780 - mae: 131.5780 - mape: 87.2176 - r2: -154.8652 - rmse: 132.0063 - val_loss: 130.5924 - val_mae: 131.0924 - val_mape: 86.9777 - val_r2: -181.4115 - val_rmse: 131.4532\n",
"Epoch 58/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 130.8940 - mae: 131.3940 - mape: 87.0950 - r2: -154.3468 - rmse: 131.8228 - val_loss: 130.4069 - val_mae: 130.9069 - val_mape: 86.8540 - val_r2: -180.8983 - val_rmse: 131.2681\n",
"Epoch 59/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 130.7074 - mae: 131.2074 - mape: 86.9707 - r2: -153.3065 - rmse: 131.6367 - val_loss: 130.2200 - val_mae: 130.7200 - val_mape: 86.7295 - val_r2: -180.3823 - val_rmse: 131.0818\n",
"Epoch 60/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 130.5216 - mae: 131.0216 - mape: 86.8469 - r2: -153.9539 - rmse: 131.4514 - val_loss: 130.0323 - val_mae: 130.5323 - val_mape: 86.6044 - val_r2: -179.8646 - val_rmse: 130.8946\n",
"Epoch 61/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 130.3329 - mae: 130.8329 - mape: 86.7210 - r2: -152.6297 - rmse: 131.2635 - val_loss: 129.8435 - val_mae: 130.3435 - val_mape: 86.4785 - val_r2: -179.3446 - val_rmse: 130.7063\n",
"Epoch 62/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 130.1447 - mae: 130.6447 - mape: 86.5955 - r2: -152.0126 - rmse: 131.0758 - val_loss: 129.6535 - val_mae: 130.1535 - val_mape: 86.3518 - val_r2: -178.8223 - val_rmse: 130.5169\n",
"Epoch 63/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 129.9547 - mae: 130.4547 - mape: 86.4690 - r2: -151.4319 - rmse: 130.8862 - val_loss: 129.4625 - val_mae: 129.9625 - val_mape: 86.2245 - val_r2: -178.2978 - val_rmse: 130.3264\n",
"Epoch 64/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 129.7637 - mae: 130.2637 - mape: 86.3417 - r2: -152.1890 - rmse: 130.6960 - val_loss: 129.2700 - val_mae: 129.7700 - val_mape: 86.0962 - val_r2: -177.7699 - val_rmse: 130.1344\n",
"Epoch 65/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 129.5713 - mae: 130.0713 - mape: 86.2133 - r2: -150.6683 - rmse: 130.5042 - val_loss: 129.0765 - val_mae: 129.5765 - val_mape: 85.9672 - val_r2: -177.2402 - val_rmse: 129.9415\n",
"Epoch 66/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 129.3767 - mae: 129.8767 - mape: 86.0835 - r2: -151.6469 - rmse: 130.3105 - val_loss: 128.8818 - val_mae: 129.3818 - val_mape: 85.8373 - val_r2: -176.7079 - val_rmse: 129.7473\n",
"Epoch 67/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 129.1829 - mae: 129.6829 - mape: 85.9546 - r2: -150.4186 - rmse: 130.1171 - val_loss: 128.6867 - val_mae: 129.1867 - val_mape: 85.7073 - val_r2: -176.1754 - val_rmse: 129.5528\n",
"Epoch 68/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 128.9887 - mae: 129.4887 - mape: 85.8250 - r2: -151.3501 - rmse: 129.9236 - val_loss: 128.4899 - val_mae: 128.9899 - val_mape: 85.5761 - val_r2: -175.6389 - val_rmse: 129.3565\n",
"Epoch 69/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 128.7915 - mae: 129.2915 - mape: 85.6938 - r2: -150.2327 - rmse: 129.7267 - val_loss: 128.2919 - val_mae: 128.7919 - val_mape: 85.4441 - val_r2: -175.1003 - val_rmse: 129.1591\n",
"Epoch 70/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 128.5936 - mae: 129.0936 - mape: 85.5618 - r2: -149.6135 - rmse: 129.5297 - val_loss: 128.0929 - val_mae: 128.5929 - val_mape: 85.3115 - val_r2: -174.5597 - val_rmse: 128.9607\n",
"Epoch 71/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 128.3947 - mae: 128.8947 - mape: 85.4293 - r2: -148.9786 - rmse: 129.3313 - val_loss: 127.8930 - val_mae: 128.3930 - val_mape: 85.1782 - val_r2: -174.0174 - val_rmse: 128.7614\n",
"Epoch 72/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 49ms/step - loss: 128.1955 - mae: 128.6955 - mape: 85.2962 - r2: -150.5439 - rmse: 129.1333 - val_loss: 127.6919 - val_mae: 128.1919 - val_mape: 85.0442 - val_r2: -173.4725 - val_rmse: 128.5608\n",
"Epoch 73/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 48ms/step - loss: 127.9942 - mae: 128.4942 - mape: 85.1620 - r2: -147.6624 - rmse: 128.9326 - val_loss: 127.4895 - val_mae: 127.9895 - val_mape: 84.9092 - val_r2: -172.9254 - val_rmse: 128.3591\n",
"Epoch 74/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 127.7916 - mae: 128.2916 - mape: 85.0270 - r2: -146.7720 - rmse: 128.7306 - val_loss: 127.2858 - val_mae: 127.7858 - val_mape: 84.7734 - val_r2: -172.3752 - val_rmse: 128.1559\n",
"Epoch 75/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 127.5894 - mae: 128.0894 - mape: 84.8923 - r2: -148.8406 - rmse: 128.5291 - val_loss: 127.0813 - val_mae: 127.5813 - val_mape: 84.6370 - val_r2: -171.8238 - val_rmse: 127.9520\n",
"Epoch 76/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 127.3845 - mae: 127.8845 - mape: 84.7559 - r2: -146.6558 - rmse: 128.3244 - val_loss: 126.8753 - val_mae: 127.3753 - val_mape: 84.4997 - val_r2: -171.2697 - val_rmse: 127.7467\n",
"Epoch 77/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 127.1788 - mae: 127.6788 - mape: 84.6186 - r2: -145.3847 - rmse: 128.1198 - val_loss: 126.6692 - val_mae: 127.1692 - val_mape: 84.3623 - val_r2: -170.7156 - val_rmse: 127.5410\n",
"Epoch 78/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 126.9727 - mae: 127.4727 - mape: 84.4813 - r2: -144.7598 - rmse: 127.9142 - val_loss: 126.4608 - val_mae: 126.9608 - val_mape: 84.2234 - val_r2: -170.1568 - val_rmse: 127.3333\n",
"Epoch 79/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 126.7648 - mae: 127.2648 - mape: 84.3427 - r2: -147.5258 - rmse: 127.7071 - val_loss: 126.2521 - val_mae: 126.7521 - val_mape: 84.0843 - val_r2: -169.5976 - val_rmse: 127.1252\n",
"Epoch 80/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 126.5569 - mae: 127.0569 - mape: 84.2039 - r2: -145.4368 - rmse: 127.5003 - val_loss: 126.0422 - val_mae: 126.5422 - val_mape: 83.9444 - val_r2: -169.0366 - val_rmse: 126.9160\n",
"Epoch 81/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 126.3474 - mae: 126.8474 - mape: 84.0645 - r2: -143.1352 - rmse: 127.2910 - val_loss: 125.8309 - val_mae: 126.3309 - val_mape: 83.8035 - val_r2: -168.4726 - val_rmse: 126.7053\n",
"Epoch 82/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 126.1355 - mae: 126.6355 - mape: 83.9232 - r2: -143.3624 - rmse: 127.0800 - val_loss: 125.6188 - val_mae: 126.1188 - val_mape: 83.6621 - val_r2: -167.9070 - val_rmse: 126.4937\n",
"Epoch 83/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 125.9247 - mae: 126.4247 - mape: 83.7826 - r2: -142.8872 - rmse: 126.8701 - val_loss: 125.4052 - val_mae: 125.9052 - val_mape: 83.5197 - val_r2: -167.3389 - val_rmse: 126.2808\n",
"Epoch 84/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 125.7120 - mae: 126.2120 - mape: 83.6410 - r2: -142.7129 - rmse: 126.6578 - val_loss: 125.1908 - val_mae: 125.6908 - val_mape: 83.3767 - val_r2: -166.7695 - val_rmse: 126.0670\n",
"Epoch 85/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 125.4970 - mae: 125.9970 - mape: 83.4974 - r2: -141.9760 - rmse: 126.4440 - val_loss: 124.9752 - val_mae: 125.4752 - val_mape: 83.2330 - val_r2: -166.1977 - val_rmse: 125.8521\n",
"Epoch 86/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 125.2826 - mae: 125.7826 - mape: 83.3546 - r2: -141.1960 - rmse: 126.2302 - val_loss: 124.7584 - val_mae: 125.2584 - val_mape: 83.0885 - val_r2: -165.6241 - val_rmse: 125.6360\n",
"Epoch 87/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 125.0649 - mae: 125.5649 - mape: 83.2095 - r2: -140.7664 - rmse: 126.0132 - val_loss: 124.5410 - val_mae: 125.0410 - val_mape: 82.9436 - val_r2: -165.0496 - val_rmse: 125.4192\n",
"Epoch 88/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 124.8480 - mae: 125.3480 - mape: 83.0651 - r2: -141.1451 - rmse: 125.7968 - val_loss: 124.3219 - val_mae: 124.8219 - val_mape: 82.7975 - val_r2: -164.4717 - val_rmse: 125.2008\n",
"Epoch 89/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 124.6286 - mae: 125.1286 - mape: 82.9187 - r2: -141.3583 - rmse: 125.5784 - val_loss: 124.1022 - val_mae: 124.6022 - val_mape: 82.6510 - val_r2: -163.8932 - val_rmse: 124.9817\n",
"Epoch 90/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 124.4112 - mae: 124.9112 - mape: 82.7737 - r2: -139.8395 - rmse: 125.3621 - val_loss: 123.8807 - val_mae: 124.3807 - val_mape: 82.5034 - val_r2: -163.3111 - val_rmse: 124.7609\n",
"Epoch 91/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 124.1902 - mae: 124.6902 - mape: 82.6264 - r2: -138.8933 - rmse: 125.1416 - val_loss: 123.6587 - val_mae: 124.1587 - val_mape: 82.3554 - val_r2: -162.7286 - val_rmse: 124.5396\n",
"Epoch 92/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 123.9690 - mae: 124.4690 - mape: 82.4789 - r2: -138.1969 - rmse: 124.9214 - val_loss: 123.4355 - val_mae: 123.9355 - val_mape: 82.2066 - val_r2: -162.1441 - val_rmse: 124.3171\n",
"Epoch 93/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 123.7447 - mae: 124.2447 - mape: 82.3292 - r2: -138.4274 - rmse: 124.6983 - val_loss: 123.2112 - val_mae: 123.7112 - val_mape: 82.0571 - val_r2: -161.5578 - val_rmse: 124.0935\n",
"Epoch 94/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 123.5217 - mae: 124.0217 - mape: 82.1812 - r2: -139.4815 - rmse: 124.4749 - val_loss: 122.9858 - val_mae: 123.4858 - val_mape: 81.9068 - val_r2: -160.9695 - val_rmse: 123.8688\n",
"Epoch 95/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 123.2968 - mae: 123.7968 - mape: 82.0309 - r2: -137.0435 - rmse: 124.2513 - val_loss: 122.7593 - val_mae: 123.2593 - val_mape: 81.7558 - val_r2: -160.3796 - val_rmse: 123.6430\n",
"Epoch 96/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 123.0705 - mae: 123.5705 - mape: 81.8799 - r2: -137.6169 - rmse: 124.0263 - val_loss: 122.5315 - val_mae: 123.0315 - val_mape: 81.6040 - val_r2: -159.7875 - val_rmse: 123.4159\n",
"Epoch 97/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 122.8436 - mae: 123.3436 - mape: 81.7289 - r2: -136.3152 - rmse: 123.7997 - val_loss: 122.3029 - val_mae: 122.8029 - val_mape: 81.4516 - val_r2: -159.1941 - val_rmse: 123.1880\n",
"Epoch 98/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 122.6150 - mae: 123.1150 - mape: 81.5767 - r2: -136.7773 - rmse: 123.5717 - val_loss: 122.0732 - val_mae: 122.5732 - val_mape: 81.2985 - val_r2: -158.5992 - val_rmse: 122.9591\n",
"Epoch 99/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 122.3851 - mae: 122.8851 - mape: 81.4230 - r2: -134.9306 - rmse: 123.3435 - val_loss: 121.8424 - val_mae: 122.3424 - val_mape: 81.1446 - val_r2: -158.0025 - val_rmse: 122.7289\n",
"Epoch 100/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 122.1553 - mae: 122.6553 - mape: 81.2704 - r2: -137.4889 - rmse: 123.1136 - val_loss: 121.6105 - val_mae: 122.1105 - val_mape: 80.9899 - val_r2: -157.4040 - val_rmse: 122.4978\n",
"Epoch 101/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 121.9240 - mae: 122.4240 - mape: 81.1157 - r2: -137.0119 - rmse: 122.8840 - val_loss: 121.3776 - val_mae: 121.8776 - val_mape: 80.8347 - val_r2: -156.8041 - val_rmse: 122.2656\n",
"Epoch 102/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 121.6915 - mae: 122.1915 - mape: 80.9607 - r2: -134.4549 - rmse: 122.6523 - val_loss: 121.1437 - val_mae: 121.6437 - val_mape: 80.6788 - val_r2: -156.2030 - val_rmse: 122.0325\n",
"Epoch 103/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 50ms/step - loss: 121.4576 - mae: 121.9576 - mape: 80.8049 - r2: -132.9466 - rmse: 122.4191 - val_loss: 120.9083 - val_mae: 121.4083 - val_mape: 80.5219 - val_r2: -155.5990 - val_rmse: 121.7979\n",
"Epoch 104/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 121.2236 - mae: 121.7236 - mape: 80.6489 - r2: -132.2487 - rmse: 122.1861 - val_loss: 120.6722 - val_mae: 121.1722 - val_mape: 80.3645 - val_r2: -154.9944 - val_rmse: 121.5625\n",
"Epoch 105/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 120.9877 - mae: 121.4877 - mape: 80.4918 - r2: -131.5870 - rmse: 121.9509 - val_loss: 120.4351 - val_mae: 120.9351 - val_mape: 80.2064 - val_r2: -154.3884 - val_rmse: 121.3261\n",
"Epoch 106/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 120.7514 - mae: 121.2514 - mape: 80.3339 - r2: -131.9983 - rmse: 121.7162 - val_loss: 120.1970 - val_mae: 120.6970 - val_mape: 80.0477 - val_r2: -153.7810 - val_rmse: 121.0888\n",
"Epoch 107/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 120.5137 - mae: 121.0137 - mape: 80.1758 - r2: -131.0121 - rmse: 121.4789 - val_loss: 119.9577 - val_mae: 120.4577 - val_mape: 79.8881 - val_r2: -153.1719 - val_rmse: 120.8503\n",
"Epoch 108/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 120.2745 - mae: 120.7745 - mape: 80.0163 - r2: -129.6228 - rmse: 121.2405 - val_loss: 119.7172 - val_mae: 120.2172 - val_mape: 79.7278 - val_r2: -152.5608 - val_rmse: 120.6105\n",
"Epoch 109/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 120.0338 - mae: 120.5338 - mape: 79.8560 - r2: -130.4699 - rmse: 121.0002 - val_loss: 119.4757 - val_mae: 119.9757 - val_mape: 79.5668 - val_r2: -151.9484 - val_rmse: 120.3698\n",
"Epoch 110/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 119.7928 - mae: 120.2928 - mape: 79.6953 - r2: -129.2888 - rmse: 120.7604 - val_loss: 119.2332 - val_mae: 119.7332 - val_mape: 79.4052 - val_r2: -151.3349 - val_rmse: 120.1282\n",
"Epoch 111/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 119.5504 - mae: 120.0504 - mape: 79.5338 - r2: -128.9569 - rmse: 120.5190 - val_loss: 118.9895 - val_mae: 119.4895 - val_mape: 79.2427 - val_r2: -150.7195 - val_rmse: 119.8853\n",
"Epoch 112/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 119.3081 - mae: 119.8081 - mape: 79.3719 - r2: -128.5213 - rmse: 120.2783 - val_loss: 118.7451 - val_mae: 119.2451 - val_mape: 79.0798 - val_r2: -150.1036 - val_rmse: 119.6417\n",
"Epoch 113/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 119.0630 - mae: 119.5630 - mape: 79.2089 - r2: -127.7393 - rmse: 120.0333 - val_loss: 118.4996 - val_mae: 118.9996 - val_mape: 78.9161 - val_r2: -149.4861 - val_rmse: 119.3969\n",
"Epoch 114/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 118.8190 - mae: 119.3190 - mape: 79.0461 - r2: -127.8189 - rmse: 119.7906 - val_loss: 118.2528 - val_mae: 118.7528 - val_mape: 78.7516 - val_r2: -148.8668 - val_rmse: 119.1510\n",
"Epoch 115/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 118.5721 - mae: 119.0721 - mape: 78.8813 - r2: -126.9974 - rmse: 119.5451 - val_loss: 118.0051 - val_mae: 118.5051 - val_mape: 78.5864 - val_r2: -148.2462 - val_rmse: 118.9041\n",
"Epoch 116/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 118.3252 - mae: 118.8252 - mape: 78.7168 - r2: -127.4985 - rmse: 119.2991 - val_loss: 117.7562 - val_mae: 118.2562 - val_mape: 78.4205 - val_r2: -147.6242 - val_rmse: 118.6561\n",
"Epoch 117/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 118.0765 - mae: 118.5765 - mape: 78.5512 - r2: -126.8991 - rmse: 119.0508 - val_loss: 117.5063 - val_mae: 118.0063 - val_mape: 78.2539 - val_r2: -147.0011 - val_rmse: 118.4071\n",
"Epoch 118/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 117.8269 - mae: 118.3269 - mape: 78.3847 - r2: -125.0494 - rmse: 118.8026 - val_loss: 117.2554 - val_mae: 117.7554 - val_mape: 78.0867 - val_r2: -146.3766 - val_rmse: 118.1570\n",
"Epoch 119/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 117.5762 - mae: 118.0762 - mape: 78.2174 - r2: -125.7715 - rmse: 118.5532 - val_loss: 117.0037 - val_mae: 117.5037 - val_mape: 77.9189 - val_r2: -145.7516 - val_rmse: 117.9062\n",
"Epoch 120/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 117.3237 - mae: 117.8237 - mape: 78.0493 - r2: -123.9535 - rmse: 118.3013 - val_loss: 116.7507 - val_mae: 117.2507 - val_mape: 77.7502 - val_r2: -145.1246 - val_rmse: 117.6540\n",
"Epoch 121/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 117.0740 - mae: 117.5740 - mape: 77.8829 - r2: -125.2403 - rmse: 118.0527 - val_loss: 116.4967 - val_mae: 116.9967 - val_mape: 77.5808 - val_r2: -144.4964 - val_rmse: 117.4008\n",
"Epoch 122/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 116.8187 - mae: 117.3187 - mape: 77.7126 - r2: -123.9459 - rmse: 117.7985 - val_loss: 116.2419 - val_mae: 116.7419 - val_mape: 77.4110 - val_r2: -143.8676 - val_rmse: 117.1469\n",
"Epoch 123/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 116.5639 - mae: 117.0639 - mape: 77.5429 - r2: -123.4611 - rmse: 117.5444 - val_loss: 115.9859 - val_mae: 116.4859 - val_mape: 77.2403 - val_r2: -143.2374 - val_rmse: 116.8918\n",
"Epoch 124/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 116.3095 - mae: 116.8095 - mape: 77.3730 - r2: -122.4719 - rmse: 117.2918 - val_loss: 115.7288 - val_mae: 116.2288 - val_mape: 77.0689 - val_r2: -142.6058 - val_rmse: 116.6356\n",
"Epoch 125/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 116.0518 - mae: 116.5518 - mape: 77.2018 - r2: -122.7772 - rmse: 117.0340 - val_loss: 115.4707 - val_mae: 115.9707 - val_mape: 76.8969 - val_r2: -141.9733 - val_rmse: 116.3784\n",
"Epoch 126/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 115.7945 - mae: 116.2945 - mape: 77.0303 - r2: -121.3219 - rmse: 116.7777 - val_loss: 115.2119 - val_mae: 115.7119 - val_mape: 76.7243 - val_r2: -141.3402 - val_rmse: 116.1205\n",
"Epoch 127/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 115.5357 - mae: 116.0357 - mape: 76.8577 - r2: -120.4905 - rmse: 116.5202 - val_loss: 114.9518 - val_mae: 115.4518 - val_mape: 76.5509 - val_r2: -140.7055 - val_rmse: 115.8613\n",
"Epoch 128/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 115.2737 - mae: 115.7737 - mape: 76.6829 - r2: -119.6875 - rmse: 116.2596 - val_loss: 114.6905 - val_mae: 115.1905 - val_mape: 76.3768 - val_r2: -140.0694 - val_rmse: 115.6010\n",
"Epoch 129/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 115.0145 - mae: 115.5145 - mape: 76.5100 - r2: -120.5758 - rmse: 116.0016 - val_loss: 114.4287 - val_mae: 114.9287 - val_mape: 76.2022 - val_r2: -139.4333 - val_rmse: 115.3400\n",
"Epoch 130/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 114.7518 - mae: 115.2518 - mape: 76.3349 - r2: -117.9841 - rmse: 115.7401 - val_loss: 114.1655 - val_mae: 114.6655 - val_mape: 76.0267 - val_r2: -138.7954 - val_rmse: 115.0778\n",
"Epoch 131/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 114.4896 - mae: 114.9896 - mape: 76.1601 - r2: -118.7698 - rmse: 115.4789 - val_loss: 113.9013 - val_mae: 114.4013 - val_mape: 75.8506 - val_r2: -138.1566 - val_rmse: 114.8146\n",
"Epoch 132/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 114.2263 - mae: 114.7263 - mape: 75.9847 - r2: -117.4594 - rmse: 115.2167 - val_loss: 113.6362 - val_mae: 114.1362 - val_mape: 75.6739 - val_r2: -137.5171 - val_rmse: 114.5504\n",
"Epoch 133/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 113.9599 - mae: 114.4599 - mape: 75.8069 - r2: -116.6060 - rmse: 114.9517 - val_loss: 113.3700 - val_mae: 113.8700 - val_mape: 75.4964 - val_r2: -136.8763 - val_rmse: 114.2852\n",
"Epoch 134/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 113.6952 - mae: 114.1952 - mape: 75.6307 - r2: -118.1602 - rmse: 114.6879 - val_loss: 113.1029 - val_mae: 113.6029 - val_mape: 75.3184 - val_r2: -136.2350 - val_rmse: 114.0191\n",
"Epoch 135/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 113.4289 - mae: 113.9289 - mape: 75.4531 - r2: -115.3682 - rmse: 114.4227 - val_loss: 112.8347 - val_mae: 113.3347 - val_mape: 75.1396 - val_r2: -135.5925 - val_rmse: 113.7519\n",
"Epoch 136/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 113.1601 - mae: 113.6601 - mape: 75.2738 - r2: -117.2677 - rmse: 114.1553 - val_loss: 112.5654 - val_mae: 113.0654 - val_mape: 74.9601 - val_r2: -134.9489 - val_rmse: 113.4836\n",
"Epoch 137/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 112.8909 - mae: 113.3909 - mape: 75.0945 - r2: -114.6084 - rmse: 113.8871 - val_loss: 112.2954 - val_mae: 112.7954 - val_mape: 74.7801 - val_r2: -134.3051 - val_rmse: 113.2145\n",
"Epoch 138/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 112.6205 - mae: 113.1205 - mape: 74.9144 - r2: -115.0930 - rmse: 113.6174 - val_loss: 112.0242 - val_mae: 112.5242 - val_mape: 74.5993 - val_r2: -133.6602 - val_rmse: 112.9444\n",
"Epoch 139/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 112.3512 - mae: 112.8512 - mape: 74.7347 - r2: -113.4350 - rmse: 113.3498 - val_loss: 111.7518 - val_mae: 112.2518 - val_mape: 74.4177 - val_r2: -133.0137 - val_rmse: 112.6730\n",
"Epoch 140/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 112.0789 - mae: 112.5789 - mape: 74.5532 - r2: -112.8636 - rmse: 113.0785 - val_loss: 111.4788 - val_mae: 111.9788 - val_mape: 74.2357 - val_r2: -132.3674 - val_rmse: 112.4010\n",
"Epoch 141/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 111.8067 - mae: 112.3067 - mape: 74.3718 - r2: -112.6902 - rmse: 112.8075 - val_loss: 111.2046 - val_mae: 111.7046 - val_mape: 74.0529 - val_r2: -131.7202 - val_rmse: 112.1279\n",
"Epoch 142/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 111.5323 - mae: 112.0323 - mape: 74.1887 - r2: -112.7054 - rmse: 112.5347 - val_loss: 110.9296 - val_mae: 111.4296 - val_mape: 73.8695 - val_r2: -131.0722 - val_rmse: 111.8538\n",
"Epoch 143/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 48ms/step - loss: 111.2570 - mae: 111.7570 - mape: 74.0057 - r2: -112.6630 - rmse: 112.2596 - val_loss: 110.6532 - val_mae: 111.1532 - val_mape: 73.6853 - val_r2: -130.4229 - val_rmse: 111.5785\n",
"Epoch 144/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 110.9813 - mae: 111.4813 - mape: 73.8217 - r2: -110.5008 - rmse: 111.9857 - val_loss: 110.3760 - val_mae: 110.8760 - val_mape: 73.5005 - val_r2: -129.7732 - val_rmse: 111.3024\n",
"Epoch 145/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 110.7049 - mae: 111.2049 - mape: 73.6374 - r2: -114.5055 - rmse: 111.7107 - val_loss: 110.0977 - val_mae: 110.5977 - val_mape: 73.3150 - val_r2: -129.1227 - val_rmse: 111.0252\n",
"Epoch 146/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 110.4270 - mae: 110.9270 - mape: 73.4520 - r2: -109.9410 - rmse: 111.4341 - val_loss: 109.8187 - val_mae: 110.3187 - val_mape: 73.1290 - val_r2: -128.4718 - val_rmse: 110.7472\n",
"Epoch 147/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 110.1488 - mae: 110.6488 - mape: 73.2667 - r2: -109.1210 - rmse: 111.1569 - val_loss: 109.5385 - val_mae: 110.0385 - val_mape: 72.9422 - val_r2: -127.8201 - val_rmse: 110.4681\n",
"Epoch 148/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 109.8674 - mae: 110.3674 - mape: 73.0794 - r2: -109.4560 - rmse: 110.8763 - val_loss: 109.2573 - val_mae: 109.7573 - val_mape: 72.7547 - val_r2: -127.1676 - val_rmse: 110.1880\n",
"Epoch 149/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 109.5872 - mae: 110.0872 - mape: 72.8926 - r2: -108.2451 - rmse: 110.5976 - val_loss: 108.9753 - val_mae: 109.4753 - val_mape: 72.5667 - val_r2: -126.5151 - val_rmse: 109.9071\n",
"Epoch 150/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 109.3050 - mae: 109.8050 - mape: 72.7045 - r2: -108.0388 - rmse: 110.3165 - val_loss: 108.6921 - val_mae: 109.1921 - val_mape: 72.3779 - val_r2: -125.8613 - val_rmse: 109.6250\n",
"Epoch 151/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 109.0225 - mae: 109.5225 - mape: 72.5156 - r2: -107.2608 - rmse: 110.0366 - val_loss: 108.4078 - val_mae: 108.9078 - val_mape: 72.1884 - val_r2: -125.2069 - val_rmse: 109.3419\n",
"Epoch 152/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 108.7394 - mae: 109.2394 - mape: 72.3271 - r2: -106.1949 - rmse: 109.7544 - val_loss: 108.1227 - val_mae: 108.6227 - val_mape: 71.9983 - val_r2: -124.5520 - val_rmse: 109.0579\n",
"Epoch 153/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 108.4542 - mae: 108.9542 - mape: 72.1370 - r2: -105.9258 - rmse: 109.4705 - val_loss: 107.8365 - val_mae: 108.3365 - val_mape: 71.8076 - val_r2: -123.8966 - val_rmse: 108.7728\n",
"Epoch 154/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 108.1682 - mae: 108.6682 - mape: 71.9468 - r2: -106.0752 - rmse: 109.1848 - val_loss: 107.5493 - val_mae: 108.0493 - val_mape: 71.6161 - val_r2: -123.2406 - val_rmse: 108.4868\n",
"Epoch 155/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 107.8816 - mae: 108.3816 - mape: 71.7554 - r2: -105.0780 - rmse: 108.9002 - val_loss: 107.2612 - val_mae: 107.7612 - val_mape: 71.4240 - val_r2: -122.5842 - val_rmse: 108.1998\n",
"Epoch 156/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 107.5945 - mae: 108.0945 - mape: 71.5641 - r2: -104.4940 - rmse: 108.6145 - val_loss: 106.9719 - val_mae: 107.4719 - val_mape: 71.2312 - val_r2: -121.9270 - val_rmse: 107.9118\n",
"Epoch 157/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 107.3058 - mae: 107.8058 - mape: 71.3713 - r2: -104.1050 - rmse: 108.3281 - val_loss: 106.6816 - val_mae: 107.1816 - val_mape: 71.0377 - val_r2: -121.2692 - val_rmse: 107.6226\n",
"Epoch 158/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 107.0160 - mae: 107.5160 - mape: 71.1779 - r2: -104.0022 - rmse: 108.0398 - val_loss: 106.3906 - val_mae: 106.8906 - val_mape: 70.8436 - val_r2: -120.6115 - val_rmse: 107.3328\n",
"Epoch 159/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 106.7253 - mae: 107.2253 - mape: 70.9846 - r2: -103.1163 - rmse: 107.7498 - val_loss: 106.0983 - val_mae: 106.5983 - val_mape: 70.6488 - val_r2: -119.9529 - val_rmse: 107.0417\n",
"Epoch 160/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 106.4340 - mae: 106.9340 - mape: 70.7906 - r2: -103.0144 - rmse: 107.4595 - val_loss: 105.8052 - val_mae: 106.3052 - val_mape: 70.4534 - val_r2: -119.2940 - val_rmse: 106.7498\n",
"Epoch 161/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 106.1407 - mae: 106.6407 - mape: 70.5945 - r2: -102.3551 - rmse: 107.1688 - val_loss: 105.5110 - val_mae: 106.0110 - val_mape: 70.2573 - val_r2: -118.6348 - val_rmse: 106.4569\n",
"Epoch 162/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 105.8470 - mae: 106.3470 - mape: 70.3991 - r2: -102.0695 - rmse: 106.8757 - val_loss: 105.2160 - val_mae: 105.7160 - val_mape: 70.0606 - val_r2: -117.9753 - val_rmse: 106.1631\n",
"Epoch 163/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 105.5524 - mae: 106.0524 - mape: 70.2027 - r2: -100.3930 - rmse: 106.5828 - val_loss: 104.9197 - val_mae: 105.4197 - val_mape: 69.8631 - val_r2: -117.3150 - val_rmse: 105.8681\n",
"Epoch 164/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 105.2559 - mae: 105.7559 - mape: 70.0048 - r2: -100.2620 - rmse: 106.2881 - val_loss: 104.6229 - val_mae: 105.1229 - val_mape: 69.6652 - val_r2: -116.6553 - val_rmse: 105.5725\n",
"Epoch 165/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 104.9616 - mae: 105.4616 - mape: 69.8093 - r2: -100.8131 - rmse: 105.9939 - val_loss: 104.3246 - val_mae: 104.8246 - val_mape: 69.4664 - val_r2: -115.9943 - val_rmse: 105.2755\n",
"Epoch 166/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 104.6633 - mae: 105.1633 - mape: 69.6101 - r2: -99.1749 - rmse: 105.6979 - val_loss: 104.0255 - val_mae: 104.5255 - val_mape: 69.2670 - val_r2: -115.3332 - val_rmse: 104.9777\n",
"Epoch 167/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 104.3650 - mae: 104.8650 - mape: 69.4109 - r2: -99.6716 - rmse: 105.4021 - val_loss: 103.7254 - val_mae: 104.2254 - val_mape: 69.0669 - val_r2: -114.6719 - val_rmse: 104.6789\n",
"Epoch 168/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 104.0652 - mae: 104.5652 - mape: 69.2115 - r2: -98.5177 - rmse: 105.1026 - val_loss: 103.4242 - val_mae: 103.9242 - val_mape: 68.8661 - val_r2: -114.0102 - val_rmse: 104.3790\n",
"Epoch 169/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 103.7629 - mae: 104.2629 - mape: 69.0100 - r2: -97.4403 - rmse: 104.8019 - val_loss: 103.1222 - val_mae: 103.6222 - val_mape: 68.6648 - val_r2: -113.3483 - val_rmse: 104.0783\n",
"Epoch 170/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 103.4628 - mae: 103.9628 - mape: 68.8099 - r2: -96.0611 - rmse: 104.5034 - val_loss: 102.8192 - val_mae: 103.3192 - val_mape: 68.4628 - val_r2: -112.6866 - val_rmse: 103.7767\n",
"Epoch 171/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 103.1604 - mae: 103.6604 - mape: 68.6083 - r2: -95.6692 - rmse: 104.2026 - val_loss: 102.5153 - val_mae: 103.0153 - val_mape: 68.2602 - val_r2: -112.0246 - val_rmse: 103.4741\n",
"Epoch 172/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 102.8567 - mae: 103.3567 - mape: 68.4059 - r2: -96.3031 - rmse: 103.9002 - val_loss: 102.2102 - val_mae: 102.7102 - val_mape: 68.0568 - val_r2: -111.3621 - val_rmse: 103.1704\n",
"Epoch 173/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 102.5535 - mae: 103.0535 - mape: 68.2038 - r2: -94.8127 - rmse: 103.5989 - val_loss: 101.9044 - val_mae: 102.4044 - val_mape: 67.8530 - val_r2: -110.6999 - val_rmse: 102.8659\n",
"Epoch 174/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 102.2476 - mae: 102.7476 - mape: 67.9998 - r2: -94.2314 - rmse: 103.2947 - val_loss: 101.5977 - val_mae: 102.0977 - val_mape: 67.6485 - val_r2: -110.0377 - val_rmse: 102.5605\n",
"Epoch 175/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 49ms/step - loss: 101.9419 - mae: 102.4419 - mape: 67.7957 - r2: -93.8276 - rmse: 102.9913 - val_loss: 101.2897 - val_mae: 101.7897 - val_mape: 67.4431 - val_r2: -109.3748 - val_rmse: 102.2540\n",
"Epoch 176/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 48ms/step - loss: 101.6340 - mae: 102.1340 - mape: 67.5907 - r2: -93.1900 - rmse: 102.6847 - val_loss: 100.9811 - val_mae: 101.4811 - val_mape: 67.2374 - val_r2: -108.7126 - val_rmse: 101.9467\n",
"Epoch 177/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 101.3252 - mae: 101.8252 - mape: 67.3851 - r2: -93.0382 - rmse: 102.3769 - val_loss: 100.6711 - val_mae: 101.1711 - val_mape: 67.0308 - val_r2: -108.0496 - val_rmse: 101.6382\n",
"Epoch 178/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 101.0180 - mae: 101.5180 - mape: 67.1803 - r2: -92.5796 - rmse: 102.0714 - val_loss: 100.3605 - val_mae: 100.8605 - val_mape: 66.8237 - val_r2: -107.3870 - val_rmse: 101.3290\n",
"Epoch 179/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 100.7067 - mae: 101.2067 - mape: 66.9728 - r2: -91.3827 - rmse: 101.7616 - val_loss: 100.0489 - val_mae: 100.5489 - val_mape: 66.6160 - val_r2: -106.7246 - val_rmse: 101.0189\n",
"Epoch 180/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 100.3951 - mae: 100.8951 - mape: 66.7648 - r2: -91.2104 - rmse: 101.4525 - val_loss: 99.7363 - val_mae: 100.2363 - val_mape: 66.4076 - val_r2: -106.0619 - val_rmse: 100.7077\n",
"Epoch 181/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 100.0840 - mae: 100.5840 - mape: 66.5577 - r2: -90.6697 - rmse: 101.1422 - val_loss: 99.4228 - val_mae: 99.9228 - val_mape: 66.1986 - val_r2: -105.3997 - val_rmse: 100.3957\n",
"Epoch 182/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 99.7711 - mae: 100.2711 - mape: 66.3489 - r2: -90.4810 - rmse: 100.8318 - val_loss: 99.1083 - val_mae: 99.6083 - val_mape: 65.9890 - val_r2: -104.7373 - val_rmse: 100.0827\n",
"Epoch 183/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 99.4568 - mae: 99.9568 - mape: 66.1396 - r2: -88.8572 - rmse: 100.5190 - val_loss: 98.7931 - val_mae: 99.2931 - val_mape: 65.7788 - val_r2: -104.0753 - val_rmse: 99.7690\n",
"Epoch 184/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 99.1418 - mae: 99.6418 - mape: 65.9291 - r2: -89.7500 - rmse: 100.2066 - val_loss: 98.4768 - val_mae: 98.9768 - val_mape: 65.5679 - val_r2: -103.4133 - val_rmse: 99.4542\n",
"Epoch 185/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 98.8261 - mae: 99.3261 - mape: 65.7191 - r2: -88.4328 - rmse: 99.8920 - val_loss: 98.1596 - val_mae: 98.6596 - val_mape: 65.3565 - val_r2: -102.7515 - val_rmse: 99.1385\n",
"Epoch 186/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 98.5090 - mae: 99.0090 - mape: 65.5077 - r2: -87.8360 - rmse: 99.5767 - val_loss: 97.8414 - val_mae: 98.3414 - val_mape: 65.1443 - val_r2: -102.0898 - val_rmse: 98.8219\n",
"Epoch 187/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 98.1927 - mae: 98.6927 - mape: 65.2969 - r2: -86.8124 - rmse: 99.2623 - val_loss: 97.5223 - val_mae: 98.0223 - val_mape: 64.9317 - val_r2: -101.4285 - val_rmse: 98.5044\n",
"Epoch 188/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 97.8738 - mae: 98.3738 - mape: 65.0844 - r2: -87.0123 - rmse: 98.9447 - val_loss: 97.2026 - val_mae: 97.7026 - val_mape: 64.7185 - val_r2: -100.7678 - val_rmse: 98.1862\n",
"Epoch 189/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 97.5545 - mae: 98.0545 - mape: 64.8709 - r2: -87.2174 - rmse: 98.6288 - val_loss: 96.8816 - val_mae: 97.3816 - val_mape: 64.5045 - val_r2: -100.1069 - val_rmse: 97.8668\n",
"Epoch 190/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 97.2344 - mae: 97.7344 - mape: 64.6582 - r2: -85.3709 - rmse: 98.3090 - val_loss: 96.5600 - val_mae: 97.0600 - val_mape: 64.2901 - val_r2: -99.4466 - val_rmse: 97.5467\n",
"Epoch 191/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 96.9119 - mae: 97.4119 - mape: 64.4429 - r2: -84.4135 - rmse: 97.9893 - val_loss: 96.2373 - val_mae: 96.7373 - val_mape: 64.0750 - val_r2: -98.7866 - val_rmse: 97.2257\n",
"Epoch 192/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 96.5908 - mae: 97.0908 - mape: 64.2288 - r2: -84.1851 - rmse: 97.6701 - val_loss: 95.9139 - val_mae: 96.4139 - val_mape: 63.8594 - val_r2: -98.1271 - val_rmse: 96.9039\n",
"Epoch 193/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 96.2684 - mae: 96.7684 - mape: 64.0143 - r2: -83.8703 - rmse: 97.3488 - val_loss: 95.5894 - val_mae: 96.0894 - val_mape: 63.6431 - val_r2: -97.4678 - val_rmse: 96.5811\n",
"Epoch 194/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 95.9430 - mae: 96.4430 - mape: 63.7972 - r2: -84.6089 - rmse: 97.0257 - val_loss: 95.2643 - val_mae: 95.7643 - val_mape: 63.4263 - val_r2: -96.8092 - val_rmse: 96.2576\n",
"Epoch 195/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 95.6174 - mae: 96.1174 - mape: 63.5804 - r2: -84.2033 - rmse: 96.7014 - val_loss: 94.9380 - val_mae: 95.4380 - val_mape: 63.2088 - val_r2: -96.1508 - val_rmse: 95.9331\n",
"Epoch 196/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 95.2933 - mae: 95.7933 - mape: 63.3640 - r2: -83.1599 - rmse: 96.3803 - val_loss: 94.6110 - val_mae: 95.1110 - val_mape: 62.9908 - val_r2: -95.4930 - val_rmse: 95.6077\n",
"Epoch 197/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 94.9662 - mae: 95.4662 - mape: 63.1463 - r2: -82.1784 - rmse: 96.0542 - val_loss: 94.2831 - val_mae: 94.7831 - val_mape: 62.7723 - val_r2: -94.8358 - val_rmse: 95.2816\n",
"Epoch 198/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 94.6391 - mae: 95.1391 - mape: 62.9277 - r2: -81.3542 - rmse: 95.7305 - val_loss: 93.9544 - val_mae: 94.4544 - val_mape: 62.5531 - val_r2: -94.1790 - val_rmse: 94.9545\n",
"Epoch 199/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 49ms/step - loss: 94.3109 - mae: 94.8109 - mape: 62.7092 - r2: -82.0628 - rmse: 95.4036 - val_loss: 93.6246 - val_mae: 94.1246 - val_mape: 62.3333 - val_r2: -93.5226 - val_rmse: 94.6265\n",
"Epoch 200/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 93.9814 - mae: 94.4814 - mape: 62.4892 - r2: -80.5396 - rmse: 95.0772 - val_loss: 93.2942 - val_mae: 93.7942 - val_mape: 62.1130 - val_r2: -92.8670 - val_rmse: 94.2978\n",
"Epoch 201/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 93.6516 - mae: 94.1516 - mape: 62.2696 - r2: -80.6569 - rmse: 94.7488 - val_loss: 92.9628 - val_mae: 93.4628 - val_mape: 61.8920 - val_r2: -92.2120 - val_rmse: 93.9682\n",
"Epoch 202/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 93.3200 - mae: 93.8200 - mape: 62.0487 - r2: -78.8336 - rmse: 94.4189 - val_loss: 92.6306 - val_mae: 93.1306 - val_mape: 61.6706 - val_r2: -91.5576 - val_rmse: 93.6378\n",
"Epoch 203/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 92.9874 - mae: 93.4874 - mape: 61.8265 - r2: -79.1243 - rmse: 94.0895 - val_loss: 92.2976 - val_mae: 92.7976 - val_mape: 61.4486 - val_r2: -90.9041 - val_rmse: 93.3066\n",
"Epoch 204/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 92.6543 - mae: 93.1543 - mape: 61.6048 - r2: -78.0827 - rmse: 93.7577 - val_loss: 91.9636 - val_mae: 92.4636 - val_mape: 61.2259 - val_r2: -90.2509 - val_rmse: 92.9745\n",
"Epoch 205/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 49ms/step - loss: 92.3217 - mae: 92.8217 - mape: 61.3834 - r2: -76.8088 - rmse: 93.4268 - val_loss: 91.6289 - val_mae: 92.1289 - val_mape: 61.0028 - val_r2: -89.5986 - val_rmse: 92.6415\n",
"Epoch 206/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 91.9865 - mae: 92.4865 - mape: 61.1598 - r2: -76.0672 - rmse: 93.0942 - val_loss: 91.2933 - val_mae: 91.7933 - val_mape: 60.7791 - val_r2: -88.9471 - val_rmse: 92.3079\n",
"Epoch 207/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 91.6516 - mae: 92.1516 - mape: 60.9365 - r2: -76.2754 - rmse: 92.7615 - val_loss: 90.9569 - val_mae: 91.4569 - val_mape: 60.5548 - val_r2: -88.2963 - val_rmse: 91.9733\n",
"Epoch 208/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 91.3159 - mae: 91.8159 - mape: 60.7128 - r2: -75.5477 - rmse: 92.4279 - val_loss: 90.6195 - val_mae: 91.1195 - val_mape: 60.3299 - val_r2: -87.6462 - val_rmse: 91.6379\n",
"Epoch 209/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 90.9787 - mae: 91.4787 - mape: 60.4880 - r2: -74.5027 - rmse: 92.0929 - val_loss: 90.2815 - val_mae: 90.7815 - val_mape: 60.1046 - val_r2: -86.9971 - val_rmse: 91.3018\n",
"Epoch 210/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 90.6402 - mae: 91.1402 - mape: 60.2620 - r2: -74.3185 - rmse: 91.7574 - val_loss: 89.9425 - val_mae: 90.4425 - val_mape: 59.8786 - val_r2: -86.3485 - val_rmse: 90.9647\n",
"Epoch 211/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 90.3022 - mae: 90.8022 - mape: 60.0371 - r2: -75.2932 - rmse: 91.4210 - val_loss: 89.6029 - val_mae: 90.1029 - val_mape: 59.6522 - val_r2: -85.7014 - val_rmse: 90.6271\n",
"Epoch 212/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 89.9623 - mae: 90.4623 - mape: 59.8103 - r2: -73.2522 - rmse: 91.0838 - val_loss: 89.2623 - val_mae: 89.7623 - val_mape: 59.4252 - val_r2: -85.0547 - val_rmse: 90.2885\n",
"Epoch 213/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 89.6211 - mae: 90.1211 - mape: 59.5829 - r2: -74.2774 - rmse: 90.7447 - val_loss: 88.9210 - val_mae: 89.4210 - val_mape: 59.1976 - val_r2: -84.4090 - val_rmse: 89.9491\n",
"Epoch 214/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 89.2797 - mae: 89.7797 - mape: 59.3555 - r2: -72.0302 - rmse: 90.4054 - val_loss: 88.5788 - val_mae: 89.0788 - val_mape: 58.9695 - val_r2: -83.7642 - val_rmse: 89.6089\n",
"Epoch 215/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 88.9369 - mae: 89.4369 - mape: 59.1275 - r2: -72.0727 - rmse: 90.0638 - val_loss: 88.2358 - val_mae: 88.7358 - val_mape: 58.7408 - val_r2: -83.1204 - val_rmse: 89.2680\n",
"Epoch 216/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 88.5940 - mae: 89.0940 - mape: 58.8985 - r2: -71.5950 - rmse: 89.7242 - val_loss: 87.8919 - val_mae: 88.3919 - val_mape: 58.5116 - val_r2: -82.4774 - val_rmse: 88.9262\n",
"Epoch 217/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 88.2504 - mae: 88.7504 - mape: 58.6691 - r2: -71.2325 - rmse: 89.3842 - val_loss: 87.5475 - val_mae: 88.0475 - val_mape: 58.2820 - val_r2: -81.8359 - val_rmse: 88.5838\n",
"Epoch 218/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 48ms/step - loss: 87.9063 - mae: 88.4063 - mape: 58.4398 - r2: -70.5790 - rmse: 89.0424 - val_loss: 87.2021 - val_mae: 87.7021 - val_mape: 58.0517 - val_r2: -81.1950 - val_rmse: 88.2405\n",
"Epoch 219/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 87.5602 - mae: 88.0602 - mape: 58.2095 - r2: -69.3876 - rmse: 88.6976 - val_loss: 86.8559 - val_mae: 87.3559 - val_mape: 57.8209 - val_r2: -80.5553 - val_rmse: 87.8964\n",
"Epoch 220/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 87.2138 - mae: 87.7138 - mape: 57.9781 - r2: -69.1578 - rmse: 88.3550 - val_loss: 86.5089 - val_mae: 87.0089 - val_mape: 57.5896 - val_r2: -79.9167 - val_rmse: 87.5516\n",
"Epoch 221/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 86.8664 - mae: 87.3664 - mape: 57.7464 - r2: -68.4424 - rmse: 88.0103 - val_loss: 86.1612 - val_mae: 86.6612 - val_mape: 57.3578 - val_r2: -79.2792 - val_rmse: 87.2061\n",
"Epoch 222/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 86.5183 - mae: 87.0183 - mape: 57.5145 - r2: -67.7217 - rmse: 87.6643 - val_loss: 85.8126 - val_mae: 86.3126 - val_mape: 57.1254 - val_r2: -78.6426 - val_rmse: 86.8596\n",
"Epoch 223/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 86.1698 - mae: 86.6698 - mape: 57.2821 - r2: -67.0855 - rmse: 87.3186 - val_loss: 85.4633 - val_mae: 85.9633 - val_mape: 56.8926 - val_r2: -78.0075 - val_rmse: 86.5126\n",
"Epoch 224/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 85.8195 - mae: 86.3195 - mape: 57.0487 - r2: -66.6095 - rmse: 86.9707 - val_loss: 85.1132 - val_mae: 85.6132 - val_mape: 56.6591 - val_r2: -77.3733 - val_rmse: 86.1647\n",
"Epoch 225/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 85.4694 - mae: 85.9694 - mape: 56.8154 - r2: -66.2413 - rmse: 86.6231 - val_loss: 84.7624 - val_mae: 85.2624 - val_mape: 56.4253 - val_r2: -76.7405 - val_rmse: 85.8161\n",
"Epoch 226/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 85.1172 - mae: 85.6172 - mape: 56.5810 - r2: -66.0857 - rmse: 86.2726 - val_loss: 84.4107 - val_mae: 84.9107 - val_mape: 56.1908 - val_r2: -76.1088 - val_rmse: 85.4667\n",
"Epoch 227/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 84.7654 - mae: 85.2654 - mape: 56.3464 - r2: -66.0088 - rmse: 85.9238 - val_loss: 84.0583 - val_mae: 84.5583 - val_mape: 55.9559 - val_r2: -75.4783 - val_rmse: 85.1166\n",
"Epoch 228/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 84.4115 - mae: 84.9115 - mape: 56.1108 - r2: -65.9496 - rmse: 85.5715 - val_loss: 83.7051 - val_mae: 84.2051 - val_mape: 55.7205 - val_r2: -74.8492 - val_rmse: 84.7658\n",
"Epoch 229/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 84.0583 - mae: 84.5583 - mape: 55.8748 - r2: -64.1267 - rmse: 85.2228 - val_loss: 83.3512 - val_mae: 83.8512 - val_mape: 55.4845 - val_r2: -74.2214 - val_rmse: 84.4142\n",
"Epoch 230/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 83.7028 - mae: 84.2028 - mape: 55.6376 - r2: -63.6403 - rmse: 84.8706 - val_loss: 82.9965 - val_mae: 83.4965 - val_mape: 55.2481 - val_r2: -73.5947 - val_rmse: 84.0619\n",
"Epoch 231/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 83.3477 - mae: 83.8477 - mape: 55.4010 - r2: -63.2279 - rmse: 84.5181 - val_loss: 82.6410 - val_mae: 83.1410 - val_mape: 55.0111 - val_r2: -72.9694 - val_rmse: 83.7088\n",
"Epoch 232/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 82.9895 - mae: 83.4895 - mape: 55.1620 - r2: -62.4607 - rmse: 84.1633 - val_loss: 82.2849 - val_mae: 82.7849 - val_mape: 54.7737 - val_r2: -72.3456 - val_rmse: 83.3551\n",
"Epoch 233/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 49ms/step - loss: 82.6348 - mae: 83.1348 - mape: 54.9263 - r2: -62.1201 - rmse: 83.8095 - val_loss: 81.9279 - val_mae: 82.4279 - val_mape: 54.5357 - val_r2: -71.7230 - val_rmse: 83.0006\n",
"Epoch 234/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 62ms/step - loss: 82.2757 - mae: 82.7757 - mape: 54.6864 - r2: -61.4099 - rmse: 83.4549 - val_loss: 81.5701 - val_mae: 82.0701 - val_mape: 54.2972 - val_r2: -71.1017 - val_rmse: 82.6453\n",
"Epoch 235/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 81.9157 - mae: 82.4157 - mape: 54.4467 - r2: -61.1632 - rmse: 83.0967 - val_loss: 81.2119 - val_mae: 81.7119 - val_mape: 54.0584 - val_r2: -70.4824 - val_rmse: 82.2896\n",
"Epoch 236/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 81.5562 - mae: 82.0562 - mape: 54.2066 - r2: -60.2707 - rmse: 82.7419 - val_loss: 80.8526 - val_mae: 81.3526 - val_mape: 53.8188 - val_r2: -69.8640 - val_rmse: 81.9328\n",
"Epoch 237/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 81.1957 - mae: 81.6957 - mape: 53.9667 - r2: -59.3570 - rmse: 82.3830 - val_loss: 80.4927 - val_mae: 80.9927 - val_mape: 53.5789 - val_r2: -69.2472 - val_rmse: 81.5755\n",
"Epoch 238/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 80.8357 - mae: 81.3357 - mape: 53.7264 - r2: -58.9882 - rmse: 82.0267 - val_loss: 80.1321 - val_mae: 80.6321 - val_mape: 53.3385 - val_r2: -68.6320 - val_rmse: 81.2175\n",
"Epoch 239/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 80.4723 - mae: 80.9723 - mape: 53.4842 - r2: -58.3461 - rmse: 81.6665 - val_loss: 79.7708 - val_mae: 80.2708 - val_mape: 53.0976 - val_r2: -68.0182 - val_rmse: 80.8587\n",
"Epoch 240/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 80.1082 - mae: 80.6082 - mape: 53.2415 - r2: -57.8946 - rmse: 81.3055 - val_loss: 79.4086 - val_mae: 79.9086 - val_mape: 52.8562 - val_r2: -67.4058 - val_rmse: 80.4992\n",
"Epoch 241/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 79.7436 - mae: 80.2436 - mape: 52.9987 - r2: -57.4216 - rmse: 80.9432 - val_loss: 79.0459 - val_mae: 79.5459 - val_mape: 52.6144 - val_r2: -66.7953 - val_rmse: 80.1392\n",
"Epoch 242/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 79.3788 - mae: 79.8788 - mape: 52.7555 - r2: -57.2431 - rmse: 80.5819 - val_loss: 78.6824 - val_mae: 79.1824 - val_mape: 52.3720 - val_r2: -66.1861 - val_rmse: 79.7783\n",
"Epoch 243/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 79.0129 - mae: 79.5129 - mape: 52.5112 - r2: -57.8018 - rmse: 80.2205 - val_loss: 78.3181 - val_mae: 78.8181 - val_mape: 52.1292 - val_r2: -65.5786 - val_rmse: 79.4168\n",
"Epoch 244/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 78.6485 - mae: 79.1485 - mape: 52.2688 - r2: -56.0245 - rmse: 79.8578 - val_loss: 77.9532 - val_mae: 78.4532 - val_mape: 51.8859 - val_r2: -64.9728 - val_rmse: 79.0547\n",
"Epoch 245/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 78.2788 - mae: 78.7788 - mape: 52.0220 - r2: -55.6164 - rmse: 79.4922 - val_loss: 77.5877 - val_mae: 78.0877 - val_mape: 51.6423 - val_r2: -64.3687 - val_rmse: 78.6919\n",
"Epoch 246/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 77.9096 - mae: 78.4096 - mape: 51.7767 - r2: -55.7301 - rmse: 79.1242 - val_loss: 77.2211 - val_mae: 77.7211 - val_mape: 51.3979 - val_r2: -63.7659 - val_rmse: 78.3282\n",
"Epoch 247/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 77.5420 - mae: 78.0420 - mape: 51.5313 - r2: -55.1348 - rmse: 78.7605 - val_loss: 76.8543 - val_mae: 77.3543 - val_mape: 51.1534 - val_r2: -63.1654 - val_rmse: 77.9642\n",
"Epoch 248/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 48ms/step - loss: 77.1725 - mae: 77.6725 - mape: 51.2848 - r2: -54.5905 - rmse: 78.3957 - val_loss: 76.4864 - val_mae: 76.9864 - val_mape: 50.9081 - val_r2: -62.5661 - val_rmse: 77.5993\n",
"Epoch 249/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 76.8015 - mae: 77.3015 - mape: 51.0374 - r2: -53.3901 - rmse: 78.0280 - val_loss: 76.1181 - val_mae: 76.6181 - val_mape: 50.6626 - val_r2: -61.9687 - val_rmse: 77.2338\n",
"Epoch 250/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 76.4288 - mae: 76.9288 - mape: 50.7889 - r2: -52.6744 - rmse: 77.6589 - val_loss: 75.7489 - val_mae: 76.2489 - val_mape: 50.4165 - val_r2: -61.3731 - val_rmse: 76.8677\n",
"Epoch 251/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 76.0560 - mae: 76.5560 - mape: 50.5406 - r2: -53.3000 - rmse: 77.2889 - val_loss: 75.3791 - val_mae: 75.8791 - val_mape: 50.1699 - val_r2: -60.7791 - val_rmse: 76.5008\n",
"Epoch 252/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 75.6840 - mae: 76.1840 - mape: 50.2928 - r2: -51.9121 - rmse: 76.9202 - val_loss: 75.0086 - val_mae: 75.5086 - val_mape: 49.9230 - val_r2: -60.1871 - val_rmse: 76.1334\n",
"Epoch 253/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 75.3112 - mae: 75.8112 - mape: 50.0438 - r2: -51.4273 - rmse: 76.5524 - val_loss: 74.6373 - val_mae: 75.1373 - val_mape: 49.6754 - val_r2: -59.5967 - val_rmse: 75.7651\n",
"Epoch 254/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 74.9369 - mae: 75.4369 - mape: 49.7941 - r2: -51.6473 - rmse: 76.1823 - val_loss: 74.2657 - val_mae: 74.7657 - val_mape: 49.4277 - val_r2: -59.0086 - val_rmse: 75.3966\n",
"Epoch 255/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 74.5593 - mae: 75.0593 - mape: 49.5428 - r2: -50.6457 - rmse: 75.8075 - val_loss: 73.8932 - val_mae: 74.3932 - val_mape: 49.1793 - val_r2: -58.4220 - val_rmse: 75.0272\n",
"Epoch 256/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 74.1833 - mae: 74.6833 - mape: 49.2918 - r2: -50.1871 - rmse: 75.4362 - val_loss: 73.5199 - val_mae: 74.0199 - val_mape: 48.9305 - val_r2: -57.8372 - val_rmse: 74.6571\n",
"Epoch 257/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 73.8047 - mae: 74.3047 - mape: 49.0393 - r2: -49.6788 - rmse: 75.0617 - val_loss: 73.1460 - val_mae: 73.6460 - val_mape: 48.6813 - val_r2: -57.2545 - val_rmse: 74.2865\n",
"Epoch 258/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 73.4301 - mae: 73.9301 - mape: 48.7900 - r2: -48.6278 - rmse: 74.6897 - val_loss: 72.7716 - val_mae: 73.2716 - val_mape: 48.4316 - val_r2: -56.6737 - val_rmse: 73.9152\n",
"Epoch 259/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 73.0499 - mae: 73.5499 - mape: 48.5366 - r2: -48.6843 - rmse: 74.3132 - val_loss: 72.3962 - val_mae: 72.8962 - val_mape: 48.1814 - val_r2: -56.0946 - val_rmse: 73.5432\n",
"Epoch 260/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 72.6703 - mae: 73.1703 - mape: 48.2835 - r2: -47.8835 - rmse: 73.9378 - val_loss: 72.0206 - val_mae: 72.5206 - val_mape: 47.9310 - val_r2: -55.5179 - val_rmse: 73.1709\n",
"Epoch 261/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 72.2911 - mae: 72.7911 - mape: 48.0308 - r2: -48.0414 - rmse: 73.5621 - val_loss: 71.6441 - val_mae: 72.1441 - val_mape: 47.6800 - val_r2: -54.9429 - val_rmse: 72.7977\n",
"Epoch 262/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 71.9111 - mae: 72.4111 - mape: 47.7770 - r2: -47.1300 - rmse: 73.1880 - val_loss: 71.2665 - val_mae: 71.7665 - val_mape: 47.4283 - val_r2: -54.3694 - val_rmse: 72.4236\n",
"Epoch 263/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 71.5291 - mae: 72.0291 - mape: 47.5227 - r2: -46.3354 - rmse: 72.8087 - val_loss: 70.8887 - val_mae: 71.3887 - val_mape: 47.1764 - val_r2: -53.7985 - val_rmse: 72.0492\n",
"Epoch 264/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 71.1479 - mae: 71.6479 - mape: 47.2685 - r2: -46.2046 - rmse: 72.4322 - val_loss: 70.5102 - val_mae: 71.0102 - val_mape: 46.9241 - val_r2: -53.2295 - val_rmse: 71.6742\n",
"Epoch 265/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 70.7672 - mae: 71.2672 - mape: 47.0143 - r2: -45.5030 - rmse: 72.0568 - val_loss: 70.1308 - val_mae: 70.6308 - val_mape: 46.6712 - val_r2: -52.6622 - val_rmse: 71.2984\n",
"Epoch 266/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 70.3821 - mae: 70.8821 - mape: 46.7579 - r2: -44.6772 - rmse: 71.6752 - val_loss: 69.7511 - val_mae: 70.2511 - val_mape: 46.4180 - val_r2: -52.0975 - val_rmse: 70.9222\n",
"Epoch 267/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 69.9992 - mae: 70.4992 - mape: 46.5032 - r2: -44.5400 - rmse: 71.2949 - val_loss: 69.3703 - val_mae: 69.8703 - val_mape: 46.1642 - val_r2: -51.5343 - val_rmse: 70.5451\n",
"Epoch 268/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 69.6130 - mae: 70.1130 - mape: 46.2454 - r2: -43.7348 - rmse: 70.9138 - val_loss: 68.9891 - val_mae: 69.4891 - val_mape: 45.9100 - val_r2: -50.9734 - val_rmse: 70.1675\n",
"Epoch 269/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 69.2289 - mae: 69.7289 - mape: 45.9892 - r2: -43.5731 - rmse: 70.5344 - val_loss: 68.6071 - val_mae: 69.1071 - val_mape: 45.6554 - val_r2: -50.4146 - val_rmse: 69.7892\n",
"Epoch 270/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 68.8435 - mae: 69.3435 - mape: 45.7324 - r2: -42.7930 - rmse: 70.1537 - val_loss: 68.2247 - val_mae: 68.7247 - val_mape: 45.4005 - val_r2: -49.8582 - val_rmse: 69.4106\n",
"Epoch 271/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 68.4568 - mae: 68.9568 - mape: 45.4753 - r2: -43.1992 - rmse: 69.7692 - val_loss: 67.8415 - val_mae: 68.3415 - val_mape: 45.1450 - val_r2: -49.3037 - val_rmse: 69.0311\n",
"Epoch 272/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 68.0707 - mae: 68.5707 - mape: 45.2175 - r2: -42.1711 - rmse: 69.3891 - val_loss: 67.4572 - val_mae: 67.9572 - val_mape: 44.8888 - val_r2: -48.7509 - val_rmse: 68.6508\n",
"Epoch 273/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 67.6836 - mae: 68.1836 - mape: 44.9595 - r2: -41.7445 - rmse: 69.0061 - val_loss: 67.0730 - val_mae: 67.5730 - val_mape: 44.6327 - val_r2: -48.2012 - val_rmse: 68.2705\n",
"Epoch 274/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 67.2936 - mae: 67.7936 - mape: 44.6993 - r2: -41.6596 - rmse: 68.6215 - val_loss: 66.6879 - val_mae: 67.1879 - val_mape: 44.3760 - val_r2: -47.6534 - val_rmse: 67.8893\n",
"Epoch 275/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 66.9057 - mae: 67.4057 - mape: 44.4408 - r2: -40.4472 - rmse: 68.2381 - val_loss: 66.3023 - val_mae: 66.8023 - val_mape: 44.1189 - val_r2: -47.1078 - val_rmse: 67.5077\n",
"Epoch 276/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 66.5166 - mae: 67.0166 - mape: 44.1811 - r2: -40.6058 - rmse: 67.8550 - val_loss: 65.9157 - val_mae: 66.4157 - val_mape: 43.8612 - val_r2: -46.5642 - val_rmse: 67.1252\n",
"Epoch 277/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 66.1259 - mae: 66.6259 - mape: 43.9209 - r2: -39.6294 - rmse: 67.4684 - val_loss: 65.5289 - val_mae: 66.0289 - val_mape: 43.6033 - val_r2: -46.0235 - val_rmse: 66.7425\n",
"Epoch 278/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 65.7342 - mae: 66.2342 - mape: 43.6601 - r2: -39.6618 - rmse: 67.0809 - val_loss: 65.1410 - val_mae: 65.6410 - val_mape: 43.3447 - val_r2: -45.4843 - val_rmse: 66.3587\n",
"Epoch 279/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 65.3439 - mae: 65.8439 - mape: 43.4001 - r2: -39.3017 - rmse: 66.6949 - val_loss: 64.7528 - val_mae: 65.2528 - val_mape: 43.0859 - val_r2: -44.9478 - val_rmse: 65.9747\n",
"Epoch 280/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 64.9529 - mae: 65.4529 - mape: 43.1392 - r2: -38.3924 - rmse: 66.3098 - val_loss: 64.3641 - val_mae: 64.8641 - val_mape: 42.8268 - val_r2: -44.4140 - val_rmse: 65.5903\n",
"Epoch 281/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 64.5625 - mae: 65.0625 - mape: 42.8785 - r2: -38.1591 - rmse: 65.9260 - val_loss: 63.9745 - val_mae: 64.4745 - val_mape: 42.5671 - val_r2: -43.8821 - val_rmse: 65.2051\n",
"Epoch 282/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 64.1685 - mae: 64.6685 - mape: 42.6162 - r2: -37.2954 - rmse: 65.5362 - val_loss: 63.5845 - val_mae: 64.0845 - val_mape: 42.3071 - val_r2: -43.3528 - val_rmse: 64.8195\n",
"Epoch 283/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 63.7740 - mae: 64.2740 - mape: 42.3534 - r2: -37.6345 - rmse: 65.1462 - val_loss: 63.1936 - val_mae: 63.6936 - val_mape: 42.0465 - val_r2: -42.8255 - val_rmse: 64.4331\n",
"Epoch 284/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 63.3807 - mae: 63.8807 - mape: 42.0909 - r2: -36.5431 - rmse: 64.7593 - val_loss: 62.8025 - val_mae: 63.3025 - val_mape: 41.7858 - val_r2: -42.3012 - val_rmse: 64.0464\n",
"Epoch 285/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 62.9862 - mae: 63.4862 - mape: 41.8281 - r2: -36.6136 - rmse: 64.3694 - val_loss: 62.4105 - val_mae: 62.9105 - val_mape: 41.5244 - val_r2: -41.7789 - val_rmse: 63.6590\n",
"Epoch 286/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 49ms/step - loss: 62.5916 - mae: 63.0916 - mape: 41.5650 - r2: -35.6594 - rmse: 63.9804 - val_loss: 62.0181 - val_mae: 62.5181 - val_mape: 41.2628 - val_r2: -41.2593 - val_rmse: 63.2712\n",
"Epoch 287/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 62.1956 - mae: 62.6956 - mape: 41.3014 - r2: -35.5448 - rmse: 63.5888 - val_loss: 61.6250 - val_mae: 62.1250 - val_mape: 41.0008 - val_r2: -40.7421 - val_rmse: 62.8829\n",
"Epoch 288/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 61.7999 - mae: 62.2999 - mape: 41.0372 - r2: -34.5892 - rmse: 63.2000 - val_loss: 61.2314 - val_mae: 61.7314 - val_mape: 40.7384 - val_r2: -40.2275 - val_rmse: 62.4941\n",
"Epoch 289/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 61.4041 - mae: 61.9041 - mape: 40.7732 - r2: -34.4239 - rmse: 62.8103 - val_loss: 60.8373 - val_mae: 61.3373 - val_mape: 40.4757 - val_r2: -39.7156 - val_rmse: 62.1048\n",
"Epoch 290/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 61.0061 - mae: 61.5061 - mape: 40.5080 - r2: -33.7795 - rmse: 62.4178 - val_loss: 60.4425 - val_mae: 60.9425 - val_mape: 40.2125 - val_r2: -39.2059 - val_rmse: 61.7149\n",
"Epoch 291/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 60.6097 - mae: 61.1097 - mape: 40.2440 - r2: -33.3344 - rmse: 62.0268 - val_loss: 60.0471 - val_mae: 60.5471 - val_mape: 39.9489 - val_r2: -38.6987 - val_rmse: 61.3244\n",
"Epoch 292/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 60.2097 - mae: 60.7097 - mape: 39.9773 - r2: -32.8462 - rmse: 61.6331 - val_loss: 59.6513 - val_mae: 60.1513 - val_mape: 39.6850 - val_r2: -38.1945 - val_rmse: 60.9338\n",
"Epoch 293/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 59.8129 - mae: 60.3129 - mape: 39.7122 - r2: -33.3131 - rmse: 61.2439 - val_loss: 59.2549 - val_mae: 59.7549 - val_mape: 39.4208 - val_r2: -37.6928 - val_rmse: 60.5424\n",
"Epoch 294/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 59.4117 - mae: 59.9117 - mape: 39.4455 - r2: -32.5348 - rmse: 60.8465 - val_loss: 58.8578 - val_mae: 59.3578 - val_mape: 39.1560 - val_r2: -37.1935 - val_rmse: 60.1505\n",
"Epoch 295/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 59.0133 - mae: 59.5133 - mape: 39.1798 - r2: -31.5229 - rmse: 60.4545 - val_loss: 58.4601 - val_mae: 58.9601 - val_mape: 38.8909 - val_r2: -36.6968 - val_rmse: 59.7582\n",
"Epoch 296/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 58.6126 - mae: 59.1126 - mape: 38.9128 - r2: -31.7440 - rmse: 60.0597 - val_loss: 58.0619 - val_mae: 58.5619 - val_mape: 38.6255 - val_r2: -36.2028 - val_rmse: 59.3653\n",
"Epoch 297/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 58.2127 - mae: 58.7127 - mape: 38.6458 - r2: -30.7672 - rmse: 59.6677 - val_loss: 57.6629 - val_mae: 58.1629 - val_mape: 38.3595 - val_r2: -35.7112 - val_rmse: 58.9718\n",
"Epoch 298/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 57.8111 - mae: 58.3111 - mape: 38.3785 - r2: -30.3311 - rmse: 59.2714 - val_loss: 57.2638 - val_mae: 57.7638 - val_mape: 38.0934 - val_r2: -35.2226 - val_rmse: 58.5781\n",
"Epoch 299/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 57.4110 - mae: 57.9110 - mape: 38.1114 - r2: -30.3829 - rmse: 58.8789 - val_loss: 56.8638 - val_mae: 57.3638 - val_mape: 37.8268 - val_r2: -34.7366 - val_rmse: 58.1838\n",
"Epoch 300/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 57.0065 - mae: 57.5065 - mape: 37.8424 - r2: -29.6669 - rmse: 58.4787 - val_loss: 56.4634 - val_mae: 56.9634 - val_mape: 37.5598 - val_r2: -34.2533 - val_rmse: 57.7890\n",
"Epoch 301/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 56.6048 - mae: 57.1048 - mape: 37.5741 - r2: -29.0981 - rmse: 58.0855 - val_loss: 56.0626 - val_mae: 56.5626 - val_mape: 37.2926 - val_r2: -33.7730 - val_rmse: 57.3940\n",
"Epoch 302/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 56.2022 - mae: 56.7022 - mape: 37.3061 - r2: -28.9122 - rmse: 57.6889 - val_loss: 55.6614 - val_mae: 56.1614 - val_mape: 37.0252 - val_r2: -33.2957 - val_rmse: 56.9987\n",
"Epoch 303/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 55.7991 - mae: 56.2991 - mape: 37.0372 - r2: -28.1913 - rmse: 57.2932 - val_loss: 55.2594 - val_mae: 55.7594 - val_mape: 36.7572 - val_r2: -32.8207 - val_rmse: 56.6026\n",
"Epoch 304/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 55.3930 - mae: 55.8930 - mape: 36.7667 - r2: -28.2134 - rmse: 56.8938 - val_loss: 54.8569 - val_mae: 55.3569 - val_mape: 36.4888 - val_r2: -32.3485 - val_rmse: 56.2061\n",
"Epoch 305/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 54.9894 - mae: 55.4894 - mape: 36.4977 - r2: -27.4716 - rmse: 56.4970 - val_loss: 54.4537 - val_mae: 54.9537 - val_mape: 36.2201 - val_r2: -31.8791 - val_rmse: 55.8091\n",
"Epoch 306/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 54.5839 - mae: 55.0839 - mape: 36.2272 - r2: -28.0192 - rmse: 56.0998 - val_loss: 54.0499 - val_mae: 54.5499 - val_mape: 35.9509 - val_r2: -31.4122 - val_rmse: 55.4115\n",
"Epoch 307/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 54.1772 - mae: 54.6772 - mape: 35.9561 - r2: -27.0065 - rmse: 55.7000 - val_loss: 53.6461 - val_mae: 54.1461 - val_mape: 35.6817 - val_r2: -30.9490 - val_rmse: 55.0141\n",
"Epoch 308/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 53.7722 - mae: 54.2722 - mape: 35.6860 - r2: -26.2374 - rmse: 55.3031 - val_loss: 53.2416 - val_mae: 53.7416 - val_mape: 35.4121 - val_r2: -30.4883 - val_rmse: 54.6160\n",
"Epoch 309/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 53.3677 - mae: 53.8677 - mape: 35.4168 - r2: -26.0430 - rmse: 54.9049 - val_loss: 52.8365 - val_mae: 53.3365 - val_mape: 35.1420 - val_r2: -30.0304 - val_rmse: 54.2174\n",
"Epoch 310/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 52.9582 - mae: 53.4582 - mape: 35.1434 - r2: -25.6169 - rmse: 54.5049 - val_loss: 52.4308 - val_mae: 52.9308 - val_mape: 34.8715 - val_r2: -29.5753 - val_rmse: 53.8184\n",
"Epoch 311/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 52.5507 - mae: 53.0507 - mape: 34.8721 - r2: -25.1029 - rmse: 54.1034 - val_loss: 52.0247 - val_mae: 52.5247 - val_mape: 34.6008 - val_r2: -29.1232 - val_rmse: 53.4190\n",
"Epoch 312/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 52.1432 - mae: 52.6432 - mape: 34.6001 - r2: -24.8709 - rmse: 53.7055 - val_loss: 51.6179 - val_mae: 52.1179 - val_mape: 34.3296 - val_r2: -28.6738 - val_rmse: 53.0190\n",
"Epoch 313/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 51.7342 - mae: 52.2342 - mape: 34.3279 - r2: -24.3171 - rmse: 53.3034 - val_loss: 51.2108 - val_mae: 51.7108 - val_mape: 34.0582 - val_r2: -28.2276 - val_rmse: 52.6189\n",
"Epoch 314/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 51.3265 - mae: 51.8265 - mape: 34.0558 - r2: -23.9997 - rmse: 52.9049 - val_loss: 50.8030 - val_mae: 51.3030 - val_mape: 33.7863 - val_r2: -27.7841 - val_rmse: 52.2181\n",
"Epoch 315/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 50.9167 - mae: 51.4167 - mape: 33.7829 - r2: -23.7700 - rmse: 52.5024 - val_loss: 50.3952 - val_mae: 50.8952 - val_mape: 33.5145 - val_r2: -27.3441 - val_rmse: 51.8175\n",
"Epoch 316/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 50.5061 - mae: 51.0061 - mape: 33.5094 - r2: -23.4041 - rmse: 52.0995 - val_loss: 49.9867 - val_mae: 50.4867 - val_mape: 33.2422 - val_r2: -26.9070 - val_rmse: 51.4164\n",
"Epoch 317/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 50.0953 - mae: 50.5953 - mape: 33.2352 - r2: -23.0374 - rmse: 51.6991 - val_loss: 49.5771 - val_mae: 50.0771 - val_mape: 32.9691 - val_r2: -26.4722 - val_rmse: 51.0143\n",
"Epoch 318/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 49.6852 - mae: 50.1852 - mape: 32.9618 - r2: -22.7847 - rmse: 51.2982 - val_loss: 49.1677 - val_mae: 49.6677 - val_mape: 32.6961 - val_r2: -26.0411 - val_rmse: 50.6125\n",
"Epoch 319/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 49.2729 - mae: 49.7729 - mape: 32.6871 - r2: -22.0742 - rmse: 50.8941 - val_loss: 48.7578 - val_mae: 49.2578 - val_mape: 32.4229 - val_r2: -25.6131 - val_rmse: 50.2103\n",
"Epoch 320/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 48.8619 - mae: 49.3619 - mape: 32.4127 - r2: -21.9799 - rmse: 50.4941 - val_loss: 48.3470 - val_mae: 48.8470 - val_mape: 32.1491 - val_r2: -25.1877 - val_rmse: 49.8074\n",
"Epoch 321/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 48.4499 - mae: 48.9499 - mape: 32.1380 - r2: -21.6426 - rmse: 50.0916 - val_loss: 47.9362 - val_mae: 48.4362 - val_mape: 31.8752 - val_r2: -24.7658 - val_rmse: 49.4046\n",
"Epoch 322/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 48.0380 - mae: 48.5380 - mape: 31.8640 - r2: -21.1362 - rmse: 49.6864 - val_loss: 47.5249 - val_mae: 48.0249 - val_mape: 31.6010 - val_r2: -24.3469 - val_rmse: 49.0013\n",
"Epoch 323/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 47.6246 - mae: 48.1246 - mape: 31.5881 - r2: -20.8258 - rmse: 49.2842 - val_loss: 47.1128 - val_mae: 47.6128 - val_mape: 31.3263 - val_r2: -23.9310 - val_rmse: 48.5976\n",
"Epoch 324/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 47.2108 - mae: 47.7108 - mape: 31.3122 - r2: -20.6426 - rmse: 48.8806 - val_loss: 46.7003 - val_mae: 47.2003 - val_mape: 31.0513 - val_r2: -23.5181 - val_rmse: 48.1935\n",
"Epoch 325/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 46.7975 - mae: 47.2975 - mape: 31.0371 - r2: -19.8993 - rmse: 48.4754 - val_loss: 46.2874 - val_mae: 46.7874 - val_mape: 30.7761 - val_r2: -23.1084 - val_rmse: 47.7892\n",
"Epoch 326/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 46.3830 - mae: 46.8830 - mape: 30.7606 - r2: -19.5410 - rmse: 48.0718 - val_loss: 45.8739 - val_mae: 46.3739 - val_mape: 30.5004 - val_r2: -22.7017 - val_rmse: 47.3844\n",
"Epoch 327/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 45.9674 - mae: 46.4674 - mape: 30.4831 - r2: -19.4893 - rmse: 47.6693 - val_loss: 45.4601 - val_mae: 45.9601 - val_mape: 30.2245 - val_r2: -22.2985 - val_rmse: 46.9795\n",
"Epoch 328/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 45.5521 - mae: 46.0521 - mape: 30.2064 - r2: -19.0831 - rmse: 47.2636 - val_loss: 45.0457 - val_mae: 45.5457 - val_mape: 29.9483 - val_r2: -21.8982 - val_rmse: 46.5742\n",
"Epoch 329/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 45.1365 - mae: 45.6365 - mape: 29.9299 - r2: -18.5538 - rmse: 46.8557 - val_loss: 44.6312 - val_mae: 45.1312 - val_mape: 29.6719 - val_r2: -21.5013 - val_rmse: 46.1689\n",
"Epoch 330/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 44.7210 - mae: 45.2209 - mape: 29.6529 - r2: -18.2251 - rmse: 46.4513 - val_loss: 44.2160 - val_mae: 44.7160 - val_mape: 29.3951 - val_r2: -21.1075 - val_rmse: 45.7631\n",
"Epoch 331/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 44.3036 - mae: 44.8036 - mape: 29.3747 - r2: -18.1202 - rmse: 46.0451 - val_loss: 43.8006 - val_mae: 44.3006 - val_mape: 29.1182 - val_r2: -20.7172 - val_rmse: 45.3573\n",
"Epoch 332/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 43.8865 - mae: 44.3865 - mape: 29.0967 - r2: -17.7665 - rmse: 45.6395 - val_loss: 43.3844 - val_mae: 43.8844 - val_mape: 28.8407 - val_r2: -20.3297 - val_rmse: 44.9509\n",
"Epoch 333/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 48ms/step - loss: 43.4703 - mae: 43.9703 - mape: 28.8192 - r2: -17.5110 - rmse: 45.2350 - val_loss: 42.9680 - val_mae: 43.4680 - val_mape: 28.5632 - val_r2: -19.9458 - val_rmse: 44.5444\n",
"Epoch 334/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 43.0527 - mae: 43.5526 - mape: 28.5405 - r2: -17.0944 - rmse: 44.8303 - val_loss: 42.5511 - val_mae: 43.0510 - val_mape: 28.2852 - val_r2: -19.5650 - val_rmse: 44.1377\n",
"Epoch 335/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 42.6334 - mae: 43.1334 - mape: 28.2613 - r2: -16.5591 - rmse: 44.4218 - val_loss: 42.1336 - val_mae: 42.6336 - val_mape: 28.0069 - val_r2: -19.1874 - val_rmse: 43.7306\n",
"Epoch 336/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 42.2134 - mae: 42.7134 - mape: 27.9813 - r2: -16.3766 - rmse: 44.0148 - val_loss: 41.7160 - val_mae: 42.2160 - val_mape: 27.7285 - val_r2: -18.8134 - val_rmse: 43.3236\n",
"Epoch 337/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 41.7963 - mae: 42.2962 - mape: 27.7033 - r2: -15.9064 - rmse: 43.6096 - val_loss: 41.2981 - val_mae: 41.7981 - val_mape: 27.4499 - val_r2: -18.4427 - val_rmse: 42.9164\n",
"Epoch 338/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 41.3774 - mae: 41.8773 - mape: 27.4243 - r2: -15.6751 - rmse: 43.2021 - val_loss: 40.8799 - val_mae: 41.3799 - val_mape: 27.1711 - val_r2: -18.0755 - val_rmse: 42.5092\n",
"Epoch 339/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 40.9573 - mae: 41.4572 - mape: 27.1443 - r2: -15.5183 - rmse: 42.7946 - val_loss: 40.4609 - val_mae: 40.9609 - val_mape: 26.8918 - val_r2: -17.7113 - val_rmse: 42.1015\n",
"Epoch 340/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 40.5382 - mae: 41.0381 - mape: 26.8648 - r2: -15.1112 - rmse: 42.3905 - val_loss: 40.0418 - val_mae: 40.5418 - val_mape: 26.6124 - val_r2: -17.3508 - val_rmse: 41.6939\n",
"Epoch 341/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 40.1191 - mae: 40.6190 - mape: 26.5858 - r2: -14.8013 - rmse: 41.9829 - val_loss: 39.6225 - val_mae: 40.1225 - val_mape: 26.3329 - val_r2: -16.9938 - val_rmse: 41.2863\n",
"Epoch 342/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 39.6984 - mae: 40.1983 - mape: 26.3058 - r2: -14.4618 - rmse: 41.5734 - val_loss: 39.2027 - val_mae: 39.7027 - val_mape: 26.0531 - val_r2: -16.6400 - val_rmse: 40.8784\n",
"Epoch 343/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 48ms/step - loss: 39.2790 - mae: 39.7789 - mape: 26.0252 - r2: -14.5762 - rmse: 41.1745 - val_loss: 38.7820 - val_mae: 39.2820 - val_mape: 25.7726 - val_r2: -16.2892 - val_rmse: 40.4700\n",
"Epoch 344/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 38.8587 - mae: 39.3585 - mape: 25.7460 - r2: -13.7789 - rmse: 40.7622 - val_loss: 38.3621 - val_mae: 38.8621 - val_mape: 25.4927 - val_r2: -15.9429 - val_rmse: 40.0626\n",
"Epoch 345/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 38.4364 - mae: 38.9362 - mape: 25.4646 - r2: -13.7206 - rmse: 40.3552 - val_loss: 37.9413 - val_mae: 38.4413 - val_mape: 25.2121 - val_r2: -15.5995 - val_rmse: 39.6545\n",
"Epoch 346/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 38.0176 - mae: 38.5175 - mape: 25.1856 - r2: -13.3033 - rmse: 39.9508 - val_loss: 37.5204 - val_mae: 38.0204 - val_mape: 24.9316 - val_r2: -15.2597 - val_rmse: 39.2466\n",
"Epoch 347/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 37.5961 - mae: 38.0960 - mape: 24.9057 - r2: -13.0208 - rmse: 39.5396 - val_loss: 37.0997 - val_mae: 37.5997 - val_mape: 24.6511 - val_r2: -14.9239 - val_rmse: 38.8392\n",
"Epoch 348/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 37.1757 - mae: 37.6757 - mape: 24.6255 - r2: -12.6513 - rmse: 39.1352 - val_loss: 36.6783 - val_mae: 37.1783 - val_mape: 24.3701 - val_r2: -14.5912 - val_rmse: 38.4313\n",
"Epoch 349/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 36.7550 - mae: 37.2550 - mape: 24.3458 - r2: -12.3770 - rmse: 38.7271 - val_loss: 36.2569 - val_mae: 36.7567 - val_mape: 24.0891 - val_r2: -14.2622 - val_rmse: 38.0237\n",
"Epoch 350/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 50ms/step - loss: 36.3340 - mae: 36.8339 - mape: 24.0654 - r2: -12.2226 - rmse: 38.3218 - val_loss: 35.8349 - val_mae: 36.3349 - val_mape: 23.8080 - val_r2: -13.9366 - val_rmse: 37.6158\n",
"Epoch 351/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 35.9123 - mae: 36.4122 - mape: 23.7845 - r2: -12.0048 - rmse: 37.9163 - val_loss: 35.4129 - val_mae: 35.9129 - val_mape: 23.5267 - val_r2: -13.6146 - val_rmse: 37.2081\n",
"Epoch 352/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 35.4927 - mae: 35.9926 - mape: 23.5051 - r2: -11.5309 - rmse: 37.5125 - val_loss: 34.9903 - val_mae: 35.4903 - val_mape: 23.2450 - val_r2: -13.2959 - val_rmse: 36.8002\n",
"Epoch 353/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 35.0698 - mae: 35.5697 - mape: 23.2239 - r2: -11.4811 - rmse: 37.1039 - val_loss: 34.5680 - val_mae: 35.0679 - val_mape: 22.9635 - val_r2: -12.9811 - val_rmse: 36.3928\n",
"Epoch 354/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 34.6483 - mae: 35.1482 - mape: 22.9434 - r2: -11.2020 - rmse: 36.6982 - val_loss: 34.1451 - val_mae: 34.6450 - val_mape: 22.6816 - val_r2: -12.6696 - val_rmse: 35.9852\n",
"Epoch 355/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 34.2267 - mae: 34.7266 - mape: 22.6628 - r2: -10.8136 - rmse: 36.2935 - val_loss: 33.7226 - val_mae: 34.2224 - val_mape: 22.4000 - val_r2: -12.3621 - val_rmse: 35.5781\n",
"Epoch 356/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 33.8060 - mae: 34.3058 - mape: 22.3830 - r2: -10.4803 - rmse: 35.8884 - val_loss: 33.2997 - val_mae: 33.7993 - val_mape: 22.1180 - val_r2: -12.0579 - val_rmse: 35.1708\n",
"Epoch 357/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 33.3837 - mae: 33.8834 - mape: 22.1020 - r2: -10.3959 - rmse: 35.4820 - val_loss: 32.8771 - val_mae: 33.3768 - val_mape: 21.8365 - val_r2: -11.7576 - val_rmse: 34.7640\n",
"Epoch 358/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 49ms/step - loss: 32.9621 - mae: 33.4618 - mape: 21.8222 - r2: -10.0606 - rmse: 35.0736 - val_loss: 32.4549 - val_mae: 32.9548 - val_mape: 21.5556 - val_r2: -11.4611 - val_rmse: 34.3577\n",
"Epoch 359/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 32.5414 - mae: 33.0412 - mape: 21.5417 - r2: -9.7963 - rmse: 34.6757 - val_loss: 32.0321 - val_mae: 32.5320 - val_mape: 21.2740 - val_r2: -11.1679 - val_rmse: 33.9510\n",
"Epoch 360/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 32.1196 - mae: 32.6195 - mape: 21.2619 - r2: -9.5612 - rmse: 34.2670 - val_loss: 31.6096 - val_mae: 32.1092 - val_mape: 20.9924 - val_r2: -10.8784 - val_rmse: 33.5447\n",
"Epoch 361/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 31.6987 - mae: 32.1984 - mape: 20.9818 - r2: -9.2226 - rmse: 33.8652 - val_loss: 31.1877 - val_mae: 31.6873 - val_mape: 20.7115 - val_r2: -10.5929 - val_rmse: 33.1392\n",
"Epoch 362/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 31.2789 - mae: 31.7786 - mape: 20.7031 - r2: -8.9984 - rmse: 33.4615 - val_loss: 30.7659 - val_mae: 31.2654 - val_mape: 20.4307 - val_r2: -10.3110 - val_rmse: 32.7337\n",
"Epoch 363/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 30.8560 - mae: 31.3558 - mape: 20.4216 - r2: -8.7912 - rmse: 33.0597 - val_loss: 30.3447 - val_mae: 30.8442 - val_mape: 20.1505 - val_r2: -10.0330 - val_rmse: 32.3290\n",
"Epoch 364/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 30.4370 - mae: 30.9368 - mape: 20.1443 - r2: -8.6926 - rmse: 32.6522 - val_loss: 29.9238 - val_mae: 30.4233 - val_mape: 19.8705 - val_r2: -9.7586 - val_rmse: 31.9244\n",
"Epoch 365/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 30.0171 - mae: 30.5169 - mape: 19.8644 - r2: -8.3463 - rmse: 32.2573 - val_loss: 29.5026 - val_mae: 30.0023 - val_mape: 19.5906 - val_r2: -9.4876 - val_rmse: 31.5198\n",
"Epoch 366/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 29.5963 - mae: 30.0961 - mape: 19.5856 - r2: -8.2189 - rmse: 31.8513 - val_loss: 29.0822 - val_mae: 29.5820 - val_mape: 19.3112 - val_r2: -9.2207 - val_rmse: 31.1161\n",
"Epoch 367/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 29.1754 - mae: 29.6751 - mape: 19.3054 - r2: -7.8698 - rmse: 31.4539 - val_loss: 28.6618 - val_mae: 29.1615 - val_mape: 19.0317 - val_r2: -8.9573 - val_rmse: 30.7125\n",
"Epoch 368/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 28.7555 - mae: 29.2552 - mape: 19.0273 - r2: -7.6236 - rmse: 31.0483 - val_loss: 28.2421 - val_mae: 28.7415 - val_mape: 18.7526 - val_r2: -8.6979 - val_rmse: 30.3098\n",
"Epoch 369/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 28.3367 - mae: 28.8364 - mape: 18.7488 - r2: -7.3600 - rmse: 30.6533 - val_loss: 27.8224 - val_mae: 28.3218 - val_mape: 18.4738 - val_r2: -8.4421 - val_rmse: 29.9074\n",
"Epoch 370/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 27.9169 - mae: 28.4165 - mape: 18.4702 - r2: -7.1524 - rmse: 30.2534 - val_loss: 27.4032 - val_mae: 27.9030 - val_mape: 18.1957 - val_r2: -8.1901 - val_rmse: 29.5057\n",
"Epoch 371/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 27.4976 - mae: 27.9974 - mape: 18.1918 - r2: -7.0447 - rmse: 29.8574 - val_loss: 26.9847 - val_mae: 27.4845 - val_mape: 17.9178 - val_r2: -7.9422 - val_rmse: 29.1050\n",
"Epoch 372/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 27.0797 - mae: 27.5795 - mape: 17.9151 - r2: -6.7276 - rmse: 29.4561 - val_loss: 26.5666 - val_mae: 27.0662 - val_mape: 17.6400 - val_r2: -7.6981 - val_rmse: 28.7050\n",
"Epoch 373/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 26.6613 - mae: 27.1609 - mape: 17.6373 - r2: -6.6071 - rmse: 29.0613 - val_loss: 26.1487 - val_mae: 26.6482 - val_mape: 17.3626 - val_r2: -7.4576 - val_rmse: 28.3054\n",
"Epoch 374/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 56ms/step - loss: 26.2455 - mae: 26.7449 - mape: 17.3618 - r2: -6.3283 - rmse: 28.6638 - val_loss: 25.7307 - val_mae: 26.2304 - val_mape: 17.0853 - val_r2: -7.2208 - val_rmse: 27.9063\n",
"Epoch 375/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 25.8289 - mae: 26.3284 - mape: 17.0861 - r2: -6.2163 - rmse: 28.2660 - val_loss: 25.3147 - val_mae: 25.8143 - val_mape: 16.8092 - val_r2: -6.9886 - val_rmse: 27.5093\n",
"Epoch 376/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 25.4127 - mae: 25.9121 - mape: 16.8094 - r2: -5.9207 - rmse: 27.8794 - val_loss: 24.8988 - val_mae: 25.3983 - val_mape: 16.5332 - val_r2: -6.7600 - val_rmse: 27.1129\n",
"Epoch 377/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 24.9969 - mae: 25.4962 - mape: 16.5341 - r2: -5.7348 - rmse: 27.4838 - val_loss: 24.4840 - val_mae: 24.9835 - val_mape: 16.2581 - val_r2: -6.5355 - val_rmse: 26.7178\n",
"Epoch 378/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 24.5835 - mae: 25.0827 - mape: 16.2604 - r2: -5.6091 - rmse: 27.0922 - val_loss: 24.0701 - val_mae: 24.5695 - val_mape: 15.9836 - val_r2: -6.3149 - val_rmse: 26.3239\n",
"Epoch 379/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 24.1711 - mae: 24.6702 - mape: 15.9875 - r2: -5.3770 - rmse: 26.7012 - val_loss: 23.6563 - val_mae: 24.1559 - val_mape: 15.7095 - val_r2: -6.0980 - val_rmse: 25.9306\n",
"Epoch 380/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 23.7608 - mae: 24.2600 - mape: 15.7156 - r2: -5.1954 - rmse: 26.3180 - val_loss: 23.2439 - val_mae: 23.7435 - val_mape: 15.4362 - val_r2: -5.8852 - val_rmse: 25.5390\n",
"Epoch 381/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 23.3484 - mae: 23.8475 - mape: 15.4427 - r2: -4.9792 - rmse: 25.9297 - val_loss: 22.8335 - val_mae: 23.3330 - val_mape: 15.1643 - val_r2: -5.6770 - val_rmse: 25.1499\n",
"Epoch 382/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 22.9411 - mae: 23.4402 - mape: 15.1734 - r2: -4.8370 - rmse: 25.5463 - val_loss: 22.4236 - val_mae: 22.9230 - val_mape: 14.8927 - val_r2: -5.4725 - val_rmse: 24.7616\n",
"Epoch 383/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 22.5364 - mae: 23.0353 - mape: 14.9063 - r2: -4.6407 - rmse: 25.1609 - val_loss: 22.0134 - val_mae: 22.5126 - val_mape: 14.6210 - val_r2: -5.2713 - val_rmse: 24.3738\n",
"Epoch 384/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 22.1313 - mae: 22.6302 - mape: 14.6390 - r2: -4.5291 - rmse: 24.7782 - val_loss: 21.6050 - val_mae: 22.1047 - val_mape: 14.3509 - val_r2: -5.0745 - val_rmse: 23.9883\n",
"Epoch 385/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 21.7253 - mae: 22.2244 - mape: 14.3701 - r2: -4.3775 - rmse: 24.4047 - val_loss: 21.1986 - val_mae: 21.6982 - val_mape: 14.0819 - val_r2: -4.8820 - val_rmse: 23.6053\n",
"Epoch 386/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 21.3238 - mae: 21.8227 - mape: 14.1059 - r2: -4.1812 - rmse: 24.0201 - val_loss: 20.7938 - val_mae: 21.2933 - val_mape: 13.8140 - val_r2: -4.6939 - val_rmse: 23.2246\n",
"Epoch 387/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 20.9229 - mae: 21.4218 - mape: 13.8406 - r2: -4.0399 - rmse: 23.6510 - val_loss: 20.3897 - val_mae: 20.8891 - val_mape: 13.5466 - val_r2: -4.5095 - val_rmse: 22.8455\n",
"Epoch 388/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 20.5290 - mae: 21.0279 - mape: 13.5813 - r2: -3.8544 - rmse: 23.2799 - val_loss: 19.9849 - val_mae: 20.4845 - val_mape: 13.2790 - val_r2: -4.3283 - val_rmse: 22.4666\n",
"Epoch 389/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 20.1289 - mae: 20.6278 - mape: 13.3177 - r2: -3.6653 - rmse: 22.9044 - val_loss: 19.5840 - val_mae: 20.0832 - val_mape: 13.0136 - val_r2: -4.1521 - val_rmse: 22.0922\n",
"Epoch 390/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 19.7355 - mae: 20.2342 - mape: 13.0585 - r2: -3.5109 - rmse: 22.5358 - val_loss: 19.1849 - val_mae: 19.6838 - val_mape: 12.7495 - val_r2: -3.9800 - val_rmse: 21.7201\n",
"Epoch 391/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 19.3437 - mae: 19.8424 - mape: 12.8006 - r2: -3.4097 - rmse: 22.1702 - val_loss: 18.7868 - val_mae: 19.2856 - val_mape: 12.4864 - val_r2: -3.8116 - val_rmse: 21.3497\n",
"Epoch 392/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 18.9550 - mae: 19.4539 - mape: 12.5448 - r2: -3.2997 - rmse: 21.8106 - val_loss: 18.3899 - val_mae: 18.8891 - val_mape: 12.2245 - val_r2: -3.6470 - val_rmse: 20.9813\n",
"Epoch 393/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 18.5658 - mae: 19.0645 - mape: 12.2891 - r2: -3.1075 - rmse: 21.4447 - val_loss: 17.9955 - val_mae: 18.4947 - val_mape: 11.9640 - val_r2: -3.4866 - val_rmse: 20.6160\n",
"Epoch 394/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 18.1784 - mae: 18.6770 - mape: 12.0340 - r2: -2.9741 - rmse: 21.0880 - val_loss: 17.6040 - val_mae: 18.1033 - val_mape: 11.7057 - val_r2: -3.3306 - val_rmse: 20.2544\n",
"Epoch 395/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 17.7957 - mae: 18.2944 - mape: 11.7832 - r2: -2.8357 - rmse: 20.7275 - val_loss: 17.2131 - val_mae: 17.7121 - val_mape: 11.4475 - val_r2: -3.1781 - val_rmse: 19.8945\n",
"Epoch 396/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 17.4172 - mae: 17.9160 - mape: 11.5344 - r2: -2.7824 - rmse: 20.3836 - val_loss: 16.8233 - val_mae: 17.3222 - val_mape: 11.1903 - val_r2: -3.0292 - val_rmse: 19.5368\n",
"Epoch 397/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 17.0339 - mae: 17.5326 - mape: 11.2830 - r2: -2.5838 - rmse: 20.0262 - val_loss: 16.4371 - val_mae: 16.9363 - val_mape: 10.9358 - val_r2: -2.8848 - val_rmse: 19.1836\n",
"Epoch 398/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 16.6569 - mae: 17.1553 - mape: 11.0353 - r2: -2.4708 - rmse: 19.6810 - val_loss: 16.0520 - val_mae: 16.5506 - val_mape: 10.6815 - val_r2: -2.7439 - val_rmse: 18.8325\n",
"Epoch 399/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 16.2761 - mae: 16.7745 - mape: 10.7848 - r2: -2.3607 - rmse: 19.3354 - val_loss: 15.6718 - val_mae: 16.1691 - val_mape: 10.4300 - val_r2: -2.6077 - val_rmse: 18.4867\n",
"Epoch 400/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 15.9094 - mae: 16.4077 - mape: 10.5399 - r2: -2.2454 - rmse: 19.0149 - val_loss: 15.3079 - val_mae: 15.8048 - val_mape: 10.1899 - val_r2: -2.4800 - val_rmse: 18.1566\n",
"Epoch 401/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 15.6218 - mae: 16.1198 - mape: 10.3565 - r2: -2.1210 - rmse: 18.7373 - val_loss: 14.9107 - val_mae: 15.4071 - val_mape: 9.9287 - val_r2: -2.3424 - val_rmse: 17.7942\n",
"Epoch 402/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 15.2819 - mae: 15.7797 - mape: 10.1345 - r2: -2.0276 - rmse: 18.4291 - val_loss: 14.5322 - val_mae: 15.0284 - val_mape: 9.6799 - val_r2: -2.2143 - val_rmse: 17.4496\n",
"Epoch 403/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 14.8892 - mae: 15.3866 - mape: 9.8772 - r2: -1.9107 - rmse: 18.0760 - val_loss: 14.1668 - val_mae: 14.6637 - val_mape: 9.4407 - val_r2: -2.0929 - val_rmse: 17.1170\n",
"Epoch 404/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 14.5078 - mae: 15.0054 - mape: 9.6284 - r2: -1.8243 - rmse: 17.7309 - val_loss: 13.8090 - val_mae: 14.3050 - val_mape: 9.2056 - val_r2: -1.9760 - val_rmse: 16.7905\n",
"Epoch 405/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 14.1364 - mae: 14.6339 - mape: 9.3862 - r2: -1.7046 - rmse: 17.3956 - val_loss: 13.4543 - val_mae: 13.9498 - val_mape: 8.9731 - val_r2: -1.8625 - val_rmse: 16.4672\n",
"Epoch 406/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 13.7758 - mae: 14.2730 - mape: 9.1511 - r2: -1.6271 - rmse: 17.0700 - val_loss: 13.1064 - val_mae: 13.6024 - val_mape: 8.7459 - val_r2: -1.7533 - val_rmse: 16.1500\n",
"Epoch 407/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 13.4205 - mae: 13.9177 - mape: 8.9198 - r2: -1.5067 - rmse: 16.7503 - val_loss: 12.7650 - val_mae: 13.2613 - val_mape: 8.5232 - val_r2: -1.6482 - val_rmse: 15.8387\n",
"Epoch 408/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 13.0721 - mae: 13.5692 - mape: 8.6925 - r2: -1.4459 - rmse: 16.4435 - val_loss: 12.4309 - val_mae: 12.9264 - val_mape: 8.3047 - val_r2: -1.5473 - val_rmse: 15.5341\n",
"Epoch 409/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 12.7348 - mae: 13.2308 - mape: 8.4721 - r2: -1.3309 - rmse: 16.1486 - val_loss: 12.1030 - val_mae: 12.5982 - val_mape: 8.0909 - val_r2: -1.4503 - val_rmse: 15.2354\n",
"Epoch 410/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 12.4089 - mae: 12.9044 - mape: 8.2605 - r2: -1.2566 - rmse: 15.8564 - val_loss: 11.7784 - val_mae: 12.2742 - val_mape: 7.8801 - val_r2: -1.3563 - val_rmse: 14.9404\n",
"Epoch 411/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 12.0827 - mae: 12.5778 - mape: 8.0487 - r2: -1.2015 - rmse: 15.5679 - val_loss: 11.4619 - val_mae: 11.9580 - val_mape: 7.6746 - val_r2: -1.2666 - val_rmse: 14.6534\n",
"Epoch 412/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 11.7682 - mae: 12.2631 - mape: 7.8443 - r2: -1.0900 - rmse: 15.3000 - val_loss: 11.1521 - val_mae: 11.6474 - val_mape: 7.4731 - val_r2: -1.1807 - val_rmse: 14.3729\n",
"Epoch 413/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 11.4631 - mae: 11.9574 - mape: 7.6469 - r2: -1.0117 - rmse: 15.0308 - val_loss: 10.8505 - val_mae: 11.3441 - val_mape: 7.2764 - val_r2: -1.0988 - val_rmse: 14.1004\n",
"Epoch 414/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 11.1631 - mae: 11.6560 - mape: 7.4520 - r2: -0.9484 - rmse: 14.7692 - val_loss: 10.5595 - val_mae: 11.0517 - val_mape: 7.0872 - val_r2: -1.0212 - val_rmse: 13.8373\n",
"Epoch 415/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 10.8742 - mae: 11.3654 - mape: 7.2645 - r2: -0.8763 - rmse: 14.5160 - val_loss: 10.2798 - val_mae: 10.7713 - val_mape: 6.9062 - val_r2: -0.9479 - val_rmse: 13.5839\n",
"Epoch 416/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 10.6030 - mae: 11.0934 - mape: 7.0904 - r2: -0.8233 - rmse: 14.2682 - val_loss: 10.0092 - val_mae: 10.4992 - val_mape: 6.7308 - val_r2: -0.8779 - val_rmse: 13.3376\n",
"Epoch 417/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 10.3339 - mae: 10.8228 - mape: 6.9162 - r2: -0.7642 - rmse: 14.0376 - val_loss: 9.7543 - val_mae: 10.2438 - val_mape: 6.5668 - val_r2: -0.8126 - val_rmse: 13.1037\n",
"Epoch 418/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 10.0877 - mae: 10.5752 - mape: 6.7581 - r2: -0.7389 - rmse: 13.8146 - val_loss: 9.5109 - val_mae: 9.9985 - val_mape: 6.4096 - val_r2: -0.7508 - val_rmse: 12.8784\n",
"Epoch 419/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 50ms/step - loss: 9.8571 - mae: 10.3440 - mape: 6.6111 - r2: -0.6884 - rmse: 13.6094 - val_loss: 9.2804 - val_mae: 9.7679 - val_mape: 6.2624 - val_r2: -0.6927 - val_rmse: 12.6631\n",
"Epoch 420/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 9.6302 - mae: 10.1167 - mape: 6.4666 - r2: -0.6103 - rmse: 13.3988 - val_loss: 9.0686 - val_mae: 9.5566 - val_mape: 6.1281 - val_r2: -0.6397 - val_rmse: 12.4631\n",
"Epoch 421/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 9.4359 - mae: 9.9213 - mape: 6.3434 - r2: -0.5934 - rmse: 13.2219 - val_loss: 8.8668 - val_mae: 9.3546 - val_mape: 6.0001 - val_r2: -0.5894 - val_rmse: 12.2704\n",
"Epoch 422/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 9.2382 - mae: 9.7228 - mape: 6.2185 - r2: -0.5171 - rmse: 13.0302 - val_loss: 8.6841 - val_mae: 9.1702 - val_mape: 5.8836 - val_r2: -0.5439 - val_rmse: 12.0937\n",
"Epoch 423/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 9.0682 - mae: 9.5517 - mape: 6.1116 - r2: -0.4810 - rmse: 12.8684 - val_loss: 8.5118 - val_mae: 8.9972 - val_mape: 5.7748 - val_r2: -0.5012 - val_rmse: 11.9251\n",
"Epoch 424/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 8.9107 - mae: 9.3939 - mape: 6.0136 - r2: -0.4711 - rmse: 12.7103 - val_loss: 8.3527 - val_mae: 8.8385 - val_mape: 5.6755 - val_r2: -0.4617 - val_rmse: 11.7673\n",
"Epoch 425/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 8.7646 - mae: 9.2475 - mape: 5.9230 - r2: -0.4030 - rmse: 12.5618 - val_loss: 8.2079 - val_mae: 8.6927 - val_mape: 5.5846 - val_r2: -0.4258 - val_rmse: 11.6219\n",
"Epoch 426/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 8.6304 - mae: 9.1132 - mape: 5.8406 - r2: -0.3880 - rmse: 12.4245 - val_loss: 8.0769 - val_mae: 8.5615 - val_mape: 5.5033 - val_r2: -0.3932 - val_rmse: 11.4884\n",
"Epoch 427/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 8.5123 - mae: 8.9955 - mape: 5.7688 - r2: -0.3606 - rmse: 12.2996 - val_loss: 7.9566 - val_mae: 8.4412 - val_mape: 5.4291 - val_r2: -0.3632 - val_rmse: 11.3636\n",
"Epoch 428/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 8.4067 - mae: 8.8900 - mape: 5.7051 - r2: -0.3540 - rmse: 12.1864 - val_loss: 7.8462 - val_mae: 8.3290 - val_mape: 5.3601 - val_r2: -0.3353 - val_rmse: 11.2468\n",
"Epoch 429/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 8.3062 - mae: 8.7892 - mape: 5.6447 - r2: -0.3006 - rmse: 12.0731 - val_loss: 7.7470 - val_mae: 8.2274 - val_mape: 5.2981 - val_r2: -0.3099 - val_rmse: 11.1395\n",
"Epoch 430/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 8.2170 - mae: 8.6997 - mape: 5.5912 - r2: -0.2823 - rmse: 11.9730 - val_loss: 7.6586 - val_mae: 8.1372 - val_mape: 5.2435 - val_r2: -0.2870 - val_rmse: 11.0415\n",
"Epoch 431/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 8.1413 - mae: 8.6231 - mape: 5.5461 - r2: -0.2577 - rmse: 11.8832 - val_loss: 7.5779 - val_mae: 8.0556 - val_mape: 5.1945 - val_r2: -0.2657 - val_rmse: 10.9497\n",
"Epoch 432/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 8.0696 - mae: 8.5508 - mape: 5.5039 - r2: -0.2395 - rmse: 11.7949 - val_loss: 7.5048 - val_mae: 7.9829 - val_mape: 5.1513 - val_r2: -0.2461 - val_rmse: 10.8646\n",
"Epoch 433/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 8.0006 - mae: 8.4819 - mape: 5.4633 - r2: -0.2208 - rmse: 11.7173 - val_loss: 7.4413 - val_mae: 7.9206 - val_mape: 5.1147 - val_r2: -0.2287 - val_rmse: 10.7888\n",
"Epoch 434/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 7.9458 - mae: 8.4274 - mape: 5.4326 - r2: -0.2052 - rmse: 11.6441 - val_loss: 7.3819 - val_mae: 7.8629 - val_mape: 5.0811 - val_r2: -0.2122 - val_rmse: 10.7160\n",
"Epoch 435/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 7.8933 - mae: 8.3754 - mape: 5.4030 - r2: -0.2011 - rmse: 11.5781 - val_loss: 7.3289 - val_mae: 7.8111 - val_mape: 5.0514 - val_r2: -0.1971 - val_rmse: 10.6492\n",
"Epoch 436/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 7.8478 - mae: 8.3304 - mape: 5.3783 - r2: -0.1813 - rmse: 11.5160 - val_loss: 7.2804 - val_mae: 7.7631 - val_mape: 5.0240 - val_r2: -0.1831 - val_rmse: 10.5864\n",
"Epoch 437/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 7.8041 - mae: 8.2873 - mape: 5.3544 - r2: -0.1744 - rmse: 11.4583 - val_loss: 7.2378 - val_mae: 7.7209 - val_mape: 5.0001 - val_r2: -0.1705 - val_rmse: 10.5300\n",
"Epoch 438/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 7.7658 - mae: 8.2496 - mape: 5.3338 - r2: -0.1600 - rmse: 11.4042 - val_loss: 7.2004 - val_mae: 7.6843 - val_mape: 4.9798 - val_r2: -0.1592 - val_rmse: 10.4790\n",
"Epoch 439/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 7.7308 - mae: 8.2154 - mape: 5.3154 - r2: -0.1558 - rmse: 11.3545 - val_loss: 7.1666 - val_mae: 7.6508 - val_mape: 4.9613 - val_r2: -0.1488 - val_rmse: 10.4321\n",
"Epoch 440/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 7.7016 - mae: 8.1868 - mape: 5.3003 - r2: -0.1488 - rmse: 11.3118 - val_loss: 7.1347 - val_mae: 7.6186 - val_mape: 4.9437 - val_r2: -0.1389 - val_rmse: 10.3869\n",
"Epoch 441/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 7.6713 - mae: 8.1562 - mape: 5.2841 - r2: -0.1296 - rmse: 11.2694 - val_loss: 7.1071 - val_mae: 7.5907 - val_mape: 4.9287 - val_r2: -0.1301 - val_rmse: 10.3468\n",
"Epoch 442/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 7.6468 - mae: 8.1318 - mape: 5.2715 - r2: -0.1224 - rmse: 11.2311 - val_loss: 7.0820 - val_mae: 7.5655 - val_mape: 4.9152 - val_r2: -0.1220 - val_rmse: 10.3098\n",
"Epoch 443/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 7.6248 - mae: 8.1098 - mape: 5.2606 - r2: -0.1250 - rmse: 11.1953 - val_loss: 7.0592 - val_mae: 7.5424 - val_mape: 4.9031 - val_r2: -0.1145 - val_rmse: 10.2753\n",
"Epoch 444/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 7.6025 - mae: 8.0878 - mape: 5.2492 - r2: -0.1128 - rmse: 11.1640 - val_loss: 7.0392 - val_mae: 7.5221 - val_mape: 4.8926 - val_r2: -0.1079 - val_rmse: 10.2445\n",
"Epoch 445/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 7.5842 - mae: 8.0698 - mape: 5.2405 - r2: -0.1152 - rmse: 11.1332 - val_loss: 7.0202 - val_mae: 7.5027 - val_mape: 4.8826 - val_r2: -0.1014 - val_rmse: 10.2144\n",
"Epoch 446/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 7.5677 - mae: 8.0534 - mape: 5.2328 - r2: -0.0982 - rmse: 11.1049 - val_loss: 7.0020 - val_mae: 7.4845 - val_mape: 4.8735 - val_r2: -0.0951 - val_rmse: 10.1851\n",
"Epoch 447/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 7.5504 - mae: 8.0363 - mape: 5.2245 - r2: -0.0984 - rmse: 11.0768 - val_loss: 6.9864 - val_mae: 7.4691 - val_mape: 4.8660 - val_r2: -0.0895 - val_rmse: 10.1594\n",
"Epoch 448/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 7.5364 - mae: 8.0226 - mape: 5.2183 - r2: -0.0938 - rmse: 11.0514 - val_loss: 6.9720 - val_mae: 7.4549 - val_mape: 4.8592 - val_r2: -0.0843 - val_rmse: 10.1349\n",
"Epoch 449/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 7.5248 - mae: 8.0111 - mape: 5.2136 - r2: -0.1152 - rmse: 11.0301 - val_loss: 6.9580 - val_mae: 7.4408 - val_mape: 4.8526 - val_r2: -0.0791 - val_rmse: 10.1107\n",
"Epoch 450/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 7.5125 - mae: 7.9993 - mape: 5.2085 - r2: -0.0816 - rmse: 11.0083 - val_loss: 6.9463 - val_mae: 7.4290 - val_mape: 4.8471 - val_r2: -0.0747 - val_rmse: 10.0899\n",
"Epoch 451/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 7.4979 - mae: 7.9847 - mape: 5.2010 - r2: -0.0849 - rmse: 10.9872 - val_loss: 6.9366 - val_mae: 7.4193 - val_mape: 4.8427 - val_r2: -0.0709 - val_rmse: 10.0723\n",
"Epoch 452/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 7.4900 - mae: 7.9771 - mape: 5.1985 - r2: -0.0743 - rmse: 10.9687 - val_loss: 6.9266 - val_mae: 7.4090 - val_mape: 4.8383 - val_r2: -0.0669 - val_rmse: 10.0532\n",
"Epoch 453/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 7.4783 - mae: 7.9654 - mape: 5.1930 - r2: -0.0699 - rmse: 10.9503 - val_loss: 6.9176 - val_mae: 7.4000 - val_mape: 4.8343 - val_r2: -0.0634 - val_rmse: 10.0366\n",
"Epoch 454/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 7.4709 - mae: 7.9579 - mape: 5.1901 - r2: -0.0693 - rmse: 10.9353 - val_loss: 6.9095 - val_mae: 7.3920 - val_mape: 4.8309 - val_r2: -0.0601 - val_rmse: 10.0210\n",
"Epoch 455/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 7.4616 - mae: 7.9487 - mape: 5.1861 - r2: -0.0777 - rmse: 10.9192 - val_loss: 6.9014 - val_mae: 7.3840 - val_mape: 4.8278 - val_r2: -0.0566 - val_rmse: 10.0044\n",
"Epoch 456/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 7.4533 - mae: 7.9405 - mape: 5.1824 - r2: -0.0644 - rmse: 10.9045 - val_loss: 6.8947 - val_mae: 7.3773 - val_mape: 4.8249 - val_r2: -0.0539 - val_rmse: 9.9917\n",
"Epoch 457/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 7.4449 - mae: 7.9321 - mape: 5.1785 - r2: -0.0689 - rmse: 10.8888 - val_loss: 6.8876 - val_mae: 7.3703 - val_mape: 4.8219 - val_r2: -0.0510 - val_rmse: 9.9779\n",
"Epoch 458/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 7.4384 - mae: 7.9255 - mape: 5.1758 - r2: -0.0597 - rmse: 10.8777 - val_loss: 6.8806 - val_mae: 7.3632 - val_mape: 4.8188 - val_r2: -0.0482 - val_rmse: 9.9648\n",
"Epoch 459/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 7.4293 - mae: 7.9164 - mape: 5.1714 - r2: -0.0491 - rmse: 10.8628 - val_loss: 6.8739 - val_mae: 7.3565 - val_mape: 4.8158 - val_r2: -0.0454 - val_rmse: 9.9516\n",
"Epoch 460/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 48ms/step - loss: 7.4206 - mae: 7.9076 - mape: 5.1664 - r2: -0.0549 - rmse: 10.8521 - val_loss: 6.8658 - val_mae: 7.3486 - val_mape: 4.8114 - val_r2: -0.0429 - val_rmse: 9.9396\n",
"Epoch 461/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 7.4117 - mae: 7.8988 - mape: 5.1610 - r2: -0.0633 - rmse: 10.8419 - val_loss: 6.8564 - val_mae: 7.3393 - val_mape: 4.8060 - val_r2: -0.0401 - val_rmse: 9.9264\n",
"Epoch 462/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 7.4064 - mae: 7.8935 - mape: 5.1599 - r2: -0.0424 - rmse: 10.8255 - val_loss: 6.8481 - val_mae: 7.3312 - val_mape: 4.8035 - val_r2: -0.0362 - val_rmse: 9.9074\n",
"Epoch 463/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 7.3882 - mae: 7.8752 - mape: 5.1465 - r2: -0.0405 - rmse: 10.8123 - val_loss: 6.8331 - val_mae: 7.3160 - val_mape: 4.7883 - val_r2: -0.0356 - val_rmse: 9.9045\n",
"Epoch 464/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 7.3867 - mae: 7.8733 - mape: 5.1365 - r2: -0.0505 - rmse: 10.8333 - val_loss: 6.8074 - val_mae: 7.2904 - val_mape: 4.7716 - val_r2: -0.0299 - val_rmse: 9.8775\n",
"Epoch 465/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 7.3572 - mae: 7.8439 - mape: 5.1261 - r2: -0.0365 - rmse: 10.7814 - val_loss: 6.8022 - val_mae: 7.2861 - val_mape: 4.7764 - val_r2: -0.0247 - val_rmse: 9.8526\n",
"Epoch 466/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 7.3467 - mae: 7.8332 - mape: 5.1103 - r2: -0.0499 - rmse: 10.8021 - val_loss: 6.7286 - val_mae: 7.2109 - val_mape: 4.7133 - val_r2: -0.0172 - val_rmse: 9.8165\n",
"Epoch 467/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 7.2638 - mae: 7.7497 - mape: 5.0579 - r2: -0.0378 - rmse: 10.7176 - val_loss: 6.6087 - val_mae: 7.0886 - val_mape: 4.6295 - val_r2: 0.0041 - val_rmse: 9.7131\n",
"Epoch 468/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 7.1237 - mae: 7.6069 - mape: 4.9376 - r2: -0.0205 - rmse: 10.6559 - val_loss: 6.2953 - val_mae: 6.7739 - val_mape: 4.4145 - val_r2: 0.0537 - val_rmse: 9.4682\n",
"Epoch 469/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 6.5297 - mae: 7.0078 - mape: 4.5049 - r2: 0.0649 - rmse: 10.2245 - val_loss: 5.2869 - val_mae: 5.7469 - val_mape: 3.6624 - val_r2: 0.1697 - val_rmse: 8.8689\n",
"Epoch 470/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 5.8890 - mae: 6.3592 - mape: 4.0406 - r2: 0.1549 - rmse: 9.6545 - val_loss: 5.0260 - val_mae: 5.4814 - val_mape: 3.4736 - val_r2: 0.2354 - val_rmse: 8.5108\n",
"Epoch 471/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 5.7288 - mae: 6.1979 - mape: 3.9279 - r2: 0.1946 - rmse: 9.4803 - val_loss: 4.8628 - val_mae: 5.3256 - val_mape: 3.3704 - val_r2: 0.2720 - val_rmse: 8.3042\n",
"Epoch 472/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 5.5509 - mae: 6.0198 - mape: 3.8101 - r2: 0.2238 - rmse: 9.3108 - val_loss: 4.6922 - val_mae: 5.1452 - val_mape: 3.2490 - val_r2: 0.2952 - val_rmse: 8.1712\n",
"Epoch 473/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 5.3175 - mae: 5.7810 - mape: 3.6452 - r2: 0.2533 - rmse: 9.1380 - val_loss: 4.6857 - val_mae: 5.1413 - val_mape: 3.2535 - val_r2: 0.2967 - val_rmse: 8.1624\n",
"Epoch 474/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 5.2020 - mae: 5.6665 - mape: 3.5754 - r2: 0.2659 - rmse: 9.0509 - val_loss: 4.5772 - val_mae: 5.0318 - val_mape: 3.1844 - val_r2: 0.3188 - val_rmse: 8.0332\n",
"Epoch 475/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 5.1099 - mae: 5.5728 - mape: 3.5106 - r2: 0.2869 - rmse: 8.9409 - val_loss: 4.5035 - val_mae: 4.9557 - val_mape: 3.1332 - val_r2: 0.3336 - val_rmse: 7.9454\n",
"Epoch 476/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 5.0760 - mae: 5.5416 - mape: 3.4960 - r2: 0.3061 - rmse: 8.8374 - val_loss: 4.6235 - val_mae: 5.0904 - val_mape: 3.2413 - val_r2: 0.3363 - val_rmse: 7.9293\n",
"Epoch 477/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 5.0382 - mae: 5.5045 - mape: 3.4750 - r2: 0.3100 - rmse: 8.7790 - val_loss: 4.4110 - val_mae: 4.8694 - val_mape: 3.0843 - val_r2: 0.3624 - val_rmse: 7.7716\n",
"Epoch 478/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 4.9287 - mae: 5.3937 - mape: 3.4006 - r2: 0.3280 - rmse: 8.6617 - val_loss: 4.3124 - val_mae: 4.7678 - val_mape: 3.0124 - val_r2: 0.3859 - val_rmse: 7.6272\n",
"Epoch 479/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 4.8025 - mae: 5.2632 - mape: 3.3179 - r2: 0.3467 - rmse: 8.5746 - val_loss: 4.2867 - val_mae: 4.7405 - val_mape: 2.9999 - val_r2: 0.3902 - val_rmse: 7.6006\n",
"Epoch 480/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 4.7121 - mae: 5.1709 - mape: 3.2586 - r2: 0.3522 - rmse: 8.4946 - val_loss: 4.2199 - val_mae: 4.6701 - val_mape: 2.9518 - val_r2: 0.3948 - val_rmse: 7.5715\n",
"Epoch 481/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 4.6603 - mae: 5.1200 - mape: 3.2254 - r2: 0.3613 - rmse: 8.4033 - val_loss: 4.1818 - val_mae: 4.6347 - val_mape: 2.9372 - val_r2: 0.4111 - val_rmse: 7.4688\n",
"Epoch 482/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 4.6037 - mae: 5.0623 - mape: 3.1917 - r2: 0.3774 - rmse: 8.3406 - val_loss: 4.0294 - val_mae: 4.4753 - val_mape: 2.8273 - val_r2: 0.4299 - val_rmse: 7.3490\n",
"Epoch 483/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step - loss: 4.5086 - mae: 4.9652 - mape: 3.1242 - r2: 0.3869 - rmse: 8.2520 - val_loss: 4.0352 - val_mae: 4.4829 - val_mape: 2.8453 - val_r2: 0.4232 - val_rmse: 7.3921\n",
"Epoch 484/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 4.4441 - mae: 4.9006 - mape: 3.0866 - r2: 0.3979 - rmse: 8.1808 - val_loss: 4.0264 - val_mae: 4.4804 - val_mape: 2.8466 - val_r2: 0.4396 - val_rmse: 7.2861\n",
"Epoch 485/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 4.3970 - mae: 4.8531 - mape: 3.0549 - r2: 0.4143 - rmse: 8.1061 - val_loss: 3.8111 - val_mae: 4.2545 - val_mape: 2.6876 - val_r2: 0.4661 - val_rmse: 7.1116\n",
"Epoch 486/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 4.3805 - mae: 4.8376 - mape: 3.0519 - r2: 0.4130 - rmse: 8.0731 - val_loss: 3.8109 - val_mae: 4.2550 - val_mape: 2.6920 - val_r2: 0.4704 - val_rmse: 7.0834\n",
"Epoch 487/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 4.2928 - mae: 4.7477 - mape: 2.9900 - r2: 0.4280 - rmse: 7.9833 - val_loss: 3.7757 - val_mae: 4.2232 - val_mape: 2.6667 - val_r2: 0.4770 - val_rmse: 7.0388\n",
"Epoch 488/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 4.2187 - mae: 4.6721 - mape: 2.9391 - r2: 0.4406 - rmse: 7.8990 - val_loss: 3.7321 - val_mae: 4.1748 - val_mape: 2.6395 - val_r2: 0.4809 - val_rmse: 7.0122\n",
"Epoch 489/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 4.1997 - mae: 4.6538 - mape: 2.9333 - r2: 0.4417 - rmse: 7.8839 - val_loss: 3.7558 - val_mae: 4.2037 - val_mape: 2.6615 - val_r2: 0.4987 - val_rmse: 6.8912\n",
"Epoch 490/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 4.1848 - mae: 4.6403 - mape: 2.9264 - r2: 0.4545 - rmse: 7.8229 - val_loss: 3.6913 - val_mae: 4.1337 - val_mape: 2.6097 - val_r2: 0.4900 - val_rmse: 6.9505\n",
"Epoch 491/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 4.1416 - mae: 4.5956 - mape: 2.8927 - r2: 0.4592 - rmse: 7.7929 - val_loss: 3.6650 - val_mae: 4.1090 - val_mape: 2.6116 - val_r2: 0.5041 - val_rmse: 6.8543\n",
"Epoch 492/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 4.0885 - mae: 4.5437 - mape: 2.8650 - r2: 0.4710 - rmse: 7.6949 - val_loss: 3.5694 - val_mae: 4.0108 - val_mape: 2.5425 - val_r2: 0.5141 - val_rmse: 6.7846\n",
"Epoch 493/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 4.0656 - mae: 4.5197 - mape: 2.8524 - r2: 0.4702 - rmse: 7.6750 - val_loss: 3.5567 - val_mae: 4.0069 - val_mape: 2.5327 - val_r2: 0.5327 - val_rmse: 6.6532\n",
"Epoch 494/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 4.0137 - mae: 4.4677 - mape: 2.8167 - r2: 0.4852 - rmse: 7.6028 - val_loss: 3.5296 - val_mae: 3.9766 - val_mape: 2.5144 - val_r2: 0.5400 - val_rmse: 6.6009\n",
"Epoch 495/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 4.0178 - mae: 4.4722 - mape: 2.8256 - r2: 0.4757 - rmse: 7.5979 - val_loss: 3.5566 - val_mae: 4.0021 - val_mape: 2.5492 - val_r2: 0.5271 - val_rmse: 6.6931\n",
"Epoch 496/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 3.9525 - mae: 4.4042 - mape: 2.7774 - r2: 0.4878 - rmse: 7.5822 - val_loss: 3.4189 - val_mae: 3.8582 - val_mape: 2.4473 - val_r2: 0.5431 - val_rmse: 6.5791\n",
"Epoch 497/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 3.9145 - mae: 4.3677 - mape: 2.7558 - r2: 0.4944 - rmse: 7.4826 - val_loss: 3.3486 - val_mae: 3.7873 - val_mape: 2.3963 - val_r2: 0.5557 - val_rmse: 6.4878\n",
"Epoch 498/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 3.8886 - mae: 4.3408 - mape: 2.7411 - r2: 0.5006 - rmse: 7.4553 - val_loss: 3.4948 - val_mae: 3.9508 - val_mape: 2.5030 - val_r2: 0.5508 - val_rmse: 6.5232\n",
"Epoch 499/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step - loss: 3.9215 - mae: 4.3765 - mape: 2.7668 - r2: 0.4948 - rmse: 7.4721 - val_loss: 3.4749 - val_mae: 3.9198 - val_mape: 2.4981 - val_r2: 0.5441 - val_rmse: 6.5717\n",
"Epoch 500/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 3.8917 - mae: 4.3465 - mape: 2.7429 - r2: 0.5045 - rmse: 7.4201 - val_loss: 3.3949 - val_mae: 3.8395 - val_mape: 2.4457 - val_r2: 0.5569 - val_rmse: 6.4788\n",
"Epoch 501/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 3.8179 - mae: 4.2712 - mape: 2.7005 - r2: 0.5235 - rmse: 7.3256 - val_loss: 3.3220 - val_mae: 3.7594 - val_mape: 2.3849 - val_r2: 0.5670 - val_rmse: 6.4045\n",
"Epoch 502/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 3.7763 - mae: 4.2262 - mape: 2.6666 - r2: 0.5251 - rmse: 7.3026 - val_loss: 3.2634 - val_mae: 3.7001 - val_mape: 2.3440 - val_r2: 0.5727 - val_rmse: 6.3620\n",
"Epoch 503/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 3.7347 - mae: 4.1862 - mape: 2.6422 - r2: 0.5208 - rmse: 7.2652 - val_loss: 3.3177 - val_mae: 3.7565 - val_mape: 2.3923 - val_r2: 0.5762 - val_rmse: 6.3360\n",
"Epoch 504/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 3.7452 - mae: 4.1965 - mape: 2.6497 - r2: 0.5259 - rmse: 7.2695 - val_loss: 3.2444 - val_mae: 3.6810 - val_mape: 2.3330 - val_r2: 0.5822 - val_rmse: 6.2912\n",
"Epoch 505/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 3.7244 - mae: 4.1759 - mape: 2.6429 - r2: 0.5314 - rmse: 7.2207 - val_loss: 3.1967 - val_mae: 3.6402 - val_mape: 2.3071 - val_r2: 0.5927 - val_rmse: 6.2119\n",
"Epoch 506/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 3.6767 - mae: 4.1282 - mape: 2.6069 - r2: 0.5468 - rmse: 7.1371 - val_loss: 3.2058 - val_mae: 3.6467 - val_mape: 2.3114 - val_r2: 0.5903 - val_rmse: 6.2301\n",
"Epoch 507/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 3.6972 - mae: 4.1493 - mape: 2.6240 - r2: 0.5392 - rmse: 7.1861 - val_loss: 3.2325 - val_mae: 3.6712 - val_mape: 2.3455 - val_r2: 0.5889 - val_rmse: 6.2407\n",
"Epoch 508/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 3.6564 - mae: 4.1093 - mape: 2.5992 - r2: 0.5467 - rmse: 7.1247 - val_loss: 3.1274 - val_mae: 3.5617 - val_mape: 2.2666 - val_r2: 0.5927 - val_rmse: 6.2119\n",
"Epoch 509/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 3.6562 - mae: 4.1069 - mape: 2.5999 - r2: 0.5440 - rmse: 7.1174 - val_loss: 3.1844 - val_mae: 3.6291 - val_mape: 2.3053 - val_r2: 0.6038 - val_rmse: 6.1264\n",
"Epoch 510/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 3.6587 - mae: 4.1113 - mape: 2.6052 - r2: 0.5489 - rmse: 7.1172 - val_loss: 3.1398 - val_mae: 3.5753 - val_mape: 2.2788 - val_r2: 0.6041 - val_rmse: 6.1239\n",
"Epoch 511/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step - loss: 3.5753 - mae: 4.0268 - mape: 2.5426 - r2: 0.5585 - rmse: 7.0286 - val_loss: 3.1398 - val_mae: 3.5784 - val_mape: 2.2807 - val_r2: 0.6025 - val_rmse: 6.1361\n",
"Epoch 512/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 3.5765 - mae: 4.0265 - mape: 2.5497 - r2: 0.5565 - rmse: 7.0373 - val_loss: 3.1025 - val_mae: 3.5492 - val_mape: 2.2570 - val_r2: 0.6182 - val_rmse: 6.0139\n",
"Epoch 513/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 3.5584 - mae: 4.0095 - mape: 2.5361 - r2: 0.5611 - rmse: 6.9611 - val_loss: 3.0708 - val_mae: 3.5109 - val_mape: 2.2363 - val_r2: 0.6108 - val_rmse: 6.0719\n",
"Epoch 514/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 3.5297 - mae: 3.9796 - mape: 2.5187 - r2: 0.5645 - rmse: 6.9545 - val_loss: 3.0074 - val_mae: 3.4418 - val_mape: 2.1877 - val_r2: 0.6190 - val_rmse: 6.0078\n",
"Epoch 515/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 3.4917 - mae: 3.9400 - mape: 2.4913 - r2: 0.5708 - rmse: 6.9452 - val_loss: 3.0075 - val_mae: 3.4393 - val_mape: 2.1949 - val_r2: 0.6147 - val_rmse: 6.0416\n",
"Epoch 516/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 3.4460 - mae: 3.8931 - mape: 2.4634 - r2: 0.5671 - rmse: 6.8895 - val_loss: 3.0069 - val_mae: 3.4394 - val_mape: 2.1892 - val_r2: 0.6168 - val_rmse: 6.0247\n",
"Epoch 517/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 3.3888 - mae: 3.8337 - mape: 2.4169 - r2: 0.5813 - rmse: 6.8280 - val_loss: 2.9269 - val_mae: 3.3564 - val_mape: 2.1356 - val_r2: 0.6246 - val_rmse: 5.9632\n",
"Epoch 518/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 3.4362 - mae: 3.8829 - mape: 2.4592 - r2: 0.5804 - rmse: 6.8685 - val_loss: 2.9864 - val_mae: 3.4182 - val_mape: 2.1840 - val_r2: 0.6225 - val_rmse: 5.9801\n",
"Epoch 519/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 47ms/step - loss: 3.3911 - mae: 3.8353 - mape: 2.4238 - r2: 0.5716 - rmse: 6.8402 - val_loss: 2.8943 - val_mae: 3.3203 - val_mape: 2.1139 - val_r2: 0.6314 - val_rmse: 5.9090\n",
"Epoch 520/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 3.3459 - mae: 3.7896 - mape: 2.3974 - r2: 0.5901 - rmse: 6.7751 - val_loss: 2.8736 - val_mae: 3.3034 - val_mape: 2.1013 - val_r2: 0.6395 - val_rmse: 5.8440\n",
"Epoch 521/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 3.3534 - mae: 3.7961 - mape: 2.4009 - r2: 0.5830 - rmse: 6.7893 - val_loss: 2.8565 - val_mae: 3.2805 - val_mape: 2.0891 - val_r2: 0.6455 - val_rmse: 5.7946\n",
"Epoch 522/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 3.3014 - mae: 3.7418 - mape: 2.3660 - r2: 0.5937 - rmse: 6.7559 - val_loss: 2.9588 - val_mae: 3.3833 - val_mape: 2.1715 - val_r2: 0.6154 - val_rmse: 6.0357\n",
"Epoch 523/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 3.2766 - mae: 3.7184 - mape: 2.3481 - r2: 0.5922 - rmse: 6.7052 - val_loss: 2.8429 - val_mae: 3.2638 - val_mape: 2.0875 - val_r2: 0.6412 - val_rmse: 5.8299\n",
"Epoch 524/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 3.2592 - mae: 3.6997 - mape: 2.3418 - r2: 0.5979 - rmse: 6.6850 - val_loss: 2.7459 - val_mae: 3.1664 - val_mape: 2.0180 - val_r2: 0.6575 - val_rmse: 5.6963\n",
"Epoch 525/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 3.2724 - mae: 3.7123 - mape: 2.3493 - r2: 0.5998 - rmse: 6.6951 - val_loss: 2.7363 - val_mae: 3.1583 - val_mape: 2.0112 - val_r2: 0.6589 - val_rmse: 5.6846\n",
"Epoch 526/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 3.2090 - mae: 3.6453 - mape: 2.3107 - r2: 0.6008 - rmse: 6.6674 - val_loss: 2.7440 - val_mae: 3.1736 - val_mape: 2.0206 - val_r2: 0.6559 - val_rmse: 5.7097\n",
"Epoch 527/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 3.2148 - mae: 3.6540 - mape: 2.3110 - r2: 0.6050 - rmse: 6.6630 - val_loss: 2.6830 - val_mae: 3.0984 - val_mape: 1.9732 - val_r2: 0.6649 - val_rmse: 5.6342\n",
"Epoch 528/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 3.1559 - mae: 3.5936 - mape: 2.2712 - r2: 0.6167 - rmse: 6.5333 - val_loss: 2.6427 - val_mae: 3.0551 - val_mape: 1.9439 - val_r2: 0.6690 - val_rmse: 5.5993\n",
"Epoch 529/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 3.1457 - mae: 3.5821 - mape: 2.2652 - r2: 0.6110 - rmse: 6.5816 - val_loss: 2.6680 - val_mae: 3.0794 - val_mape: 1.9649 - val_r2: 0.6651 - val_rmse: 5.6323\n",
"Epoch 530/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 51ms/step - loss: 3.1039 - mae: 3.5400 - mape: 2.2379 - r2: 0.6202 - rmse: 6.5194 - val_loss: 2.6244 - val_mae: 3.0356 - val_mape: 1.9297 - val_r2: 0.6740 - val_rmse: 5.5573\n",
"Epoch 531/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 3.0561 - mae: 3.4907 - mape: 2.2028 - r2: 0.6293 - rmse: 6.4464 - val_loss: 2.5804 - val_mae: 2.9996 - val_mape: 1.9065 - val_r2: 0.6800 - val_rmse: 5.5060\n",
"Epoch 532/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 3.0728 - mae: 3.5069 - mape: 2.2166 - r2: 0.6240 - rmse: 6.4804 - val_loss: 2.5725 - val_mae: 2.9829 - val_mape: 1.9017 - val_r2: 0.6795 - val_rmse: 5.5102\n",
"Epoch 533/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step - loss: 3.0249 - mae: 3.4589 - mape: 2.1837 - r2: 0.6334 - rmse: 6.4191 - val_loss: 2.6690 - val_mae: 3.0810 - val_mape: 1.9734 - val_r2: 0.6648 - val_rmse: 5.6354\n",
"Epoch 534/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 3.0514 - mae: 3.4881 - mape: 2.2061 - r2: 0.6308 - rmse: 6.4243 - val_loss: 2.6087 - val_mae: 3.0228 - val_mape: 1.9286 - val_r2: 0.6748 - val_rmse: 5.5501\n",
"Epoch 535/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 3.0253 - mae: 3.4579 - mape: 2.1837 - r2: 0.6297 - rmse: 6.4237 - val_loss: 2.5365 - val_mae: 2.9465 - val_mape: 1.8812 - val_r2: 0.6850 - val_rmse: 5.4625\n",
"Epoch 536/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 3.0360 - mae: 3.4693 - mape: 2.2019 - r2: 0.6265 - rmse: 6.4445 - val_loss: 2.5040 - val_mae: 2.9184 - val_mape: 1.8585 - val_r2: 0.6957 - val_rmse: 5.3693\n",
"Epoch 537/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 3.0152 - mae: 3.4495 - mape: 2.1797 - r2: 0.6351 - rmse: 6.4009 - val_loss: 2.6447 - val_mae: 3.0629 - val_mape: 1.9677 - val_r2: 0.6783 - val_rmse: 5.5202\n",
"Epoch 538/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 48ms/step - loss: 2.9849 - mae: 3.4205 - mape: 2.1710 - r2: 0.6393 - rmse: 6.3538 - val_loss: 2.4903 - val_mae: 2.9037 - val_mape: 1.8495 - val_r2: 0.6882 - val_rmse: 5.4352\n",
"Epoch 539/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 2.9757 - mae: 3.4109 - mape: 2.1573 - r2: 0.6380 - rmse: 6.3559 - val_loss: 2.5376 - val_mae: 2.9458 - val_mape: 1.8893 - val_r2: 0.6768 - val_rmse: 5.5335\n",
"Epoch 540/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step - loss: 2.9518 - mae: 3.3860 - mape: 2.1455 - r2: 0.6432 - rmse: 6.2887 - val_loss: 2.5045 - val_mae: 2.9144 - val_mape: 1.8611 - val_r2: 0.6944 - val_rmse: 5.3805\n",
"Epoch 541/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 2.9448 - mae: 3.3787 - mape: 2.1355 - r2: 0.6463 - rmse: 6.2869 - val_loss: 2.4998 - val_mae: 2.9125 - val_mape: 1.8627 - val_r2: 0.6885 - val_rmse: 5.4323\n",
"Epoch 542/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 2.9221 - mae: 3.3540 - mape: 2.1248 - r2: 0.6539 - rmse: 6.2360 - val_loss: 2.5069 - val_mae: 2.9162 - val_mape: 1.8701 - val_r2: 0.6883 - val_rmse: 5.4338\n",
"Epoch 543/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 2.8946 - mae: 3.3280 - mape: 2.1027 - r2: 0.6539 - rmse: 6.2093 - val_loss: 2.3949 - val_mae: 2.8046 - val_mape: 1.7887 - val_r2: 0.7045 - val_rmse: 5.2905\n",
"Epoch 544/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 2.9072 - mae: 3.3410 - mape: 2.1170 - r2: 0.6478 - rmse: 6.2146 - val_loss: 2.4695 - val_mae: 2.8773 - val_mape: 1.8445 - val_r2: 0.6920 - val_rmse: 5.4017\n",
"Epoch 545/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 2.8881 - mae: 3.3220 - mape: 2.1059 - r2: 0.6610 - rmse: 6.1730 - val_loss: 2.4605 - val_mae: 2.8883 - val_mape: 1.8447 - val_r2: 0.7032 - val_rmse: 5.3025\n",
"Epoch 546/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 2.8636 - mae: 3.2986 - mape: 2.0846 - r2: 0.6672 - rmse: 6.1043 - val_loss: 2.4211 - val_mae: 2.8376 - val_mape: 1.8204 - val_r2: 0.6995 - val_rmse: 5.3356\n",
"Epoch 547/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 2.8571 - mae: 3.2922 - mape: 2.0865 - r2: 0.6648 - rmse: 6.1252 - val_loss: 2.4155 - val_mae: 2.8279 - val_mape: 1.8060 - val_r2: 0.7088 - val_rmse: 5.2524\n",
"Epoch 548/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 2.8391 - mae: 3.2711 - mape: 2.0658 - r2: 0.6662 - rmse: 6.1184 - val_loss: 2.4144 - val_mae: 2.8185 - val_mape: 1.8129 - val_r2: 0.6967 - val_rmse: 5.3603\n",
"Epoch 549/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 2.7845 - mae: 3.2138 - mape: 2.0326 - r2: 0.6701 - rmse: 6.0539 - val_loss: 2.4080 - val_mae: 2.8183 - val_mape: 1.8048 - val_r2: 0.7116 - val_rmse: 5.2266\n",
"Epoch 550/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 2.8140 - mae: 3.2454 - mape: 2.0552 - r2: 0.6638 - rmse: 6.1125 - val_loss: 2.3909 - val_mae: 2.8001 - val_mape: 1.8007 - val_r2: 0.7074 - val_rmse: 5.2648\n",
"Epoch 551/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 2.7921 - mae: 3.2224 - mape: 2.0397 - r2: 0.6727 - rmse: 6.0506 - val_loss: 2.3889 - val_mae: 2.7995 - val_mape: 1.8009 - val_r2: 0.7103 - val_rmse: 5.2383\n",
"Epoch 552/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 2.7989 - mae: 3.2295 - mape: 2.0472 - r2: 0.6690 - rmse: 6.0951 - val_loss: 2.3877 - val_mae: 2.7947 - val_mape: 1.7915 - val_r2: 0.7067 - val_rmse: 5.2712\n",
"Epoch 553/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 2.7709 - mae: 3.1999 - mape: 2.0339 - r2: 0.6683 - rmse: 6.0998 - val_loss: 2.3301 - val_mae: 2.7394 - val_mape: 1.7556 - val_r2: 0.7246 - val_rmse: 5.1074\n",
"Epoch 554/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 2.7446 - mae: 3.1741 - mape: 2.0093 - r2: 0.6788 - rmse: 5.9944 - val_loss: 2.2934 - val_mae: 2.6965 - val_mape: 1.7284 - val_r2: 0.7205 - val_rmse: 5.1457\n",
"Epoch 555/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 2.7361 - mae: 3.1648 - mape: 2.0103 - r2: 0.6737 - rmse: 6.0358 - val_loss: 2.2202 - val_mae: 2.6220 - val_mape: 1.6769 - val_r2: 0.7350 - val_rmse: 5.0103\n",
"Epoch 556/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 2.7199 - mae: 3.1498 - mape: 1.9931 - r2: 0.6816 - rmse: 5.9639 - val_loss: 2.3713 - val_mae: 2.7784 - val_mape: 1.7882 - val_r2: 0.7159 - val_rmse: 5.1881\n",
"Epoch 557/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 2.7311 - mae: 3.1607 - mape: 2.0058 - r2: 0.6743 - rmse: 6.0224 - val_loss: 2.3634 - val_mae: 2.7722 - val_mape: 1.7848 - val_r2: 0.7166 - val_rmse: 5.1812\n",
"Epoch 558/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 2.7237 - mae: 3.1550 - mape: 1.9998 - r2: 0.6821 - rmse: 5.9433 - val_loss: 2.2300 - val_mae: 2.6377 - val_mape: 1.6932 - val_r2: 0.7403 - val_rmse: 4.9602\n",
"Epoch 559/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 2.7337 - mae: 3.1638 - mape: 2.0099 - r2: 0.6776 - rmse: 5.9668 - val_loss: 2.2874 - val_mae: 2.6945 - val_mape: 1.7269 - val_r2: 0.7222 - val_rmse: 5.1301\n",
"Epoch 560/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 2.6510 - mae: 3.0786 - mape: 1.9482 - r2: 0.6925 - rmse: 5.8611 - val_loss: 2.3437 - val_mae: 2.7592 - val_mape: 1.7778 - val_r2: 0.7191 - val_rmse: 5.1584\n",
"Epoch 561/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 2.6700 - mae: 3.0982 - mape: 1.9667 - r2: 0.6874 - rmse: 5.8842 - val_loss: 2.2598 - val_mae: 2.6690 - val_mape: 1.7166 - val_r2: 0.7343 - val_rmse: 5.0167\n",
"Epoch 562/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 2.6610 - mae: 3.0879 - mape: 1.9595 - r2: 0.6822 - rmse: 5.9129 - val_loss: 2.1990 - val_mae: 2.5999 - val_mape: 1.6691 - val_r2: 0.7325 - val_rmse: 5.0343\n",
"Epoch 563/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 2.6426 - mae: 3.0697 - mape: 1.9467 - r2: 0.6949 - rmse: 5.8488 - val_loss: 2.1967 - val_mae: 2.5980 - val_mape: 1.6689 - val_r2: 0.7415 - val_rmse: 4.9486\n",
"Epoch 564/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 2.6266 - mae: 3.0547 - mape: 1.9384 - r2: 0.6960 - rmse: 5.8305 - val_loss: 2.2409 - val_mae: 2.6476 - val_mape: 1.7065 - val_r2: 0.7266 - val_rmse: 5.0890\n",
"Epoch 565/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 2.6261 - mae: 3.0541 - mape: 1.9379 - r2: 0.6945 - rmse: 5.8593 - val_loss: 2.2359 - val_mae: 2.6439 - val_mape: 1.7040 - val_r2: 0.7313 - val_rmse: 5.0455\n",
"Epoch 566/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 52ms/step - loss: 2.6345 - mae: 3.0624 - mape: 1.9479 - r2: 0.6930 - rmse: 5.8630 - val_loss: 2.1531 - val_mae: 2.5652 - val_mape: 1.6446 - val_r2: 0.7500 - val_rmse: 4.8662\n",
"Epoch 567/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 2.5955 - mae: 3.0217 - mape: 1.9144 - r2: 0.7023 - rmse: 5.7709 - val_loss: 2.1471 - val_mae: 2.5601 - val_mape: 1.6442 - val_r2: 0.7485 - val_rmse: 4.8814\n",
"Epoch 568/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 2.5647 - mae: 2.9894 - mape: 1.8972 - r2: 0.7011 - rmse: 5.7469 - val_loss: 2.1744 - val_mae: 2.5822 - val_mape: 1.6617 - val_r2: 0.7445 - val_rmse: 4.9201\n",
"Epoch 569/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 2.5699 - mae: 2.9947 - mape: 1.9010 - r2: 0.6986 - rmse: 5.8131 - val_loss: 2.1028 - val_mae: 2.5046 - val_mape: 1.6087 - val_r2: 0.7499 - val_rmse: 4.8672\n",
"Epoch 570/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step - loss: 2.5561 - mae: 2.9807 - mape: 1.8937 - r2: 0.7047 - rmse: 5.7392 - val_loss: 2.1239 - val_mae: 2.5270 - val_mape: 1.6228 - val_r2: 0.7489 - val_rmse: 4.8767\n",
"Epoch 571/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 2.5535 - mae: 2.9793 - mape: 1.8942 - r2: 0.7040 - rmse: 5.7608 - val_loss: 2.1482 - val_mae: 2.5496 - val_mape: 1.6449 - val_r2: 0.7432 - val_rmse: 4.9318\n",
"Epoch 572/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 2.5283 - mae: 2.9549 - mape: 1.8732 - r2: 0.7090 - rmse: 5.7226 - val_loss: 2.0784 - val_mae: 2.4786 - val_mape: 1.5916 - val_r2: 0.7654 - val_rmse: 4.7144\n",
"Epoch 573/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 2.5321 - mae: 2.9567 - mape: 1.8817 - r2: 0.7065 - rmse: 5.7223 - val_loss: 2.1042 - val_mae: 2.5083 - val_mape: 1.6124 - val_r2: 0.7534 - val_rmse: 4.8335\n",
"Epoch 574/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 2.4946 - mae: 2.9176 - mape: 1.8523 - r2: 0.7084 - rmse: 5.7102 - val_loss: 2.0402 - val_mae: 2.4432 - val_mape: 1.5724 - val_r2: 0.7650 - val_rmse: 4.7186\n",
"Epoch 575/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 2.4872 - mae: 2.9108 - mape: 1.8462 - r2: 0.7169 - rmse: 5.6273 - val_loss: 2.1019 - val_mae: 2.5017 - val_mape: 1.6087 - val_r2: 0.7567 - val_rmse: 4.8009\n",
"Epoch 576/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 2.5227 - mae: 2.9471 - mape: 1.8756 - r2: 0.7103 - rmse: 5.6770 - val_loss: 2.1003 - val_mae: 2.5015 - val_mape: 1.6131 - val_r2: 0.7539 - val_rmse: 4.8280\n",
"Epoch 577/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 2.5054 - mae: 2.9297 - mape: 1.8644 - r2: 0.7091 - rmse: 5.7124 - val_loss: 2.0294 - val_mae: 2.4236 - val_mape: 1.5629 - val_r2: 0.7566 - val_rmse: 4.8022\n",
"Epoch 578/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 2.4328 - mae: 2.8545 - mape: 1.8143 - r2: 0.7158 - rmse: 5.5946 - val_loss: 2.0310 - val_mae: 2.4375 - val_mape: 1.5604 - val_r2: 0.7552 - val_rmse: 4.8154\n",
"Epoch 579/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 2.4416 - mae: 2.8651 - mape: 1.8168 - r2: 0.7198 - rmse: 5.5958 - val_loss: 2.0536 - val_mae: 2.4570 - val_mape: 1.5876 - val_r2: 0.7538 - val_rmse: 4.8289\n",
"Epoch 580/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 2.4570 - mae: 2.8799 - mape: 1.8313 - r2: 0.7210 - rmse: 5.5789 - val_loss: 2.0395 - val_mae: 2.4510 - val_mape: 1.5779 - val_r2: 0.7628 - val_rmse: 4.7402\n",
"Epoch 581/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 2.4313 - mae: 2.8533 - mape: 1.8120 - r2: 0.7184 - rmse: 5.5848 - val_loss: 1.9660 - val_mae: 2.3720 - val_mape: 1.5248 - val_r2: 0.7719 - val_rmse: 4.6481\n",
"Epoch 582/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 2.4164 - mae: 2.8386 - mape: 1.8075 - r2: 0.7243 - rmse: 5.5406 - val_loss: 2.0344 - val_mae: 2.4387 - val_mape: 1.5748 - val_r2: 0.7545 - val_rmse: 4.8220\n",
"Epoch 583/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 2.3854 - mae: 2.8074 - mape: 1.7814 - r2: 0.7238 - rmse: 5.5398 - val_loss: 1.9274 - val_mae: 2.3251 - val_mape: 1.4945 - val_r2: 0.7840 - val_rmse: 4.5239\n",
"Epoch 584/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 2.4105 - mae: 2.8323 - mape: 1.8005 - r2: 0.7255 - rmse: 5.5531 - val_loss: 1.9469 - val_mae: 2.3426 - val_mape: 1.5089 - val_r2: 0.7760 - val_rmse: 4.6067\n",
"Epoch 585/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 2.4304 - mae: 2.8521 - mape: 1.8155 - r2: 0.7231 - rmse: 5.5414 - val_loss: 1.9226 - val_mae: 2.3149 - val_mape: 1.4916 - val_r2: 0.7771 - val_rmse: 4.5952\n",
"Epoch 586/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 2.3995 - mae: 2.8222 - mape: 1.7943 - r2: 0.7212 - rmse: 5.5790 - val_loss: 1.9429 - val_mae: 2.3369 - val_mape: 1.4955 - val_r2: 0.7769 - val_rmse: 4.5968\n",
"Epoch 587/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 2.3953 - mae: 2.8173 - mape: 1.7930 - r2: 0.7284 - rmse: 5.5225 - val_loss: 2.1124 - val_mae: 2.5205 - val_mape: 1.6378 - val_r2: 0.7589 - val_rmse: 4.7793\n",
"Epoch 588/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 2.3716 - mae: 2.7922 - mape: 1.7777 - r2: 0.7285 - rmse: 5.5099 - val_loss: 2.0130 - val_mae: 2.4216 - val_mape: 1.5656 - val_r2: 0.7719 - val_rmse: 4.6488\n",
"Epoch 589/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 2.3761 - mae: 2.7992 - mape: 1.7790 - r2: 0.7256 - rmse: 5.4991 - val_loss: 2.0830 - val_mae: 2.4999 - val_mape: 1.6209 - val_r2: 0.7592 - val_rmse: 4.7763\n",
"Epoch 590/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 2.3991 - mae: 2.8217 - mape: 1.8028 - r2: 0.7276 - rmse: 5.5072 - val_loss: 1.8833 - val_mae: 2.2750 - val_mape: 1.4651 - val_r2: 0.7804 - val_rmse: 4.5607\n",
"Epoch 591/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 2.3412 - mae: 2.7576 - mape: 1.7539 - r2: 0.7336 - rmse: 5.4711 - val_loss: 1.9031 - val_mae: 2.2987 - val_mape: 1.4843 - val_r2: 0.7826 - val_rmse: 4.5382\n",
"Epoch 592/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 2.3469 - mae: 2.7668 - mape: 1.7622 - r2: 0.7370 - rmse: 5.4424 - val_loss: 1.9301 - val_mae: 2.3191 - val_mape: 1.4957 - val_r2: 0.7710 - val_rmse: 4.6574\n",
"Epoch 593/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 2.3629 - mae: 2.7833 - mape: 1.7765 - r2: 0.7284 - rmse: 5.5190 - val_loss: 1.8611 - val_mae: 2.2599 - val_mape: 1.4529 - val_r2: 0.7843 - val_rmse: 4.5206\n",
"Epoch 594/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 2.3382 - mae: 2.7614 - mape: 1.7592 - r2: 0.7373 - rmse: 5.4210 - val_loss: 1.9106 - val_mae: 2.3244 - val_mape: 1.4986 - val_r2: 0.7878 - val_rmse: 4.4838\n",
"Epoch 595/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step - loss: 2.3177 - mae: 2.7384 - mape: 1.7446 - r2: 0.7359 - rmse: 5.4405 - val_loss: 1.9046 - val_mae: 2.3039 - val_mape: 1.4820 - val_r2: 0.7779 - val_rmse: 4.5870\n",
"Epoch 596/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 2.3183 - mae: 2.7376 - mape: 1.7464 - r2: 0.7344 - rmse: 5.4516 - val_loss: 1.8913 - val_mae: 2.2925 - val_mape: 1.4856 - val_r2: 0.7801 - val_rmse: 4.5640\n",
"Epoch 597/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 2.3025 - mae: 2.7218 - mape: 1.7357 - r2: 0.7461 - rmse: 5.3197 - val_loss: 1.8849 - val_mae: 2.2874 - val_mape: 1.4821 - val_r2: 0.7858 - val_rmse: 4.5042\n",
"Epoch 598/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 2.3393 - mae: 2.7626 - mape: 1.7645 - r2: 0.7368 - rmse: 5.4305 - val_loss: 1.8484 - val_mae: 2.2617 - val_mape: 1.4522 - val_r2: 0.7997 - val_rmse: 4.3556\n",
"Epoch 599/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 2.3082 - mae: 2.7294 - mape: 1.7403 - r2: 0.7410 - rmse: 5.3666 - val_loss: 1.9364 - val_mae: 2.3408 - val_mape: 1.5176 - val_r2: 0.7805 - val_rmse: 4.5603\n",
"Epoch 600/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 2.2681 - mae: 2.6860 - mape: 1.7126 - r2: 0.7437 - rmse: 5.3622 - val_loss: 1.8346 - val_mae: 2.2355 - val_mape: 1.4363 - val_r2: 0.7822 - val_rmse: 4.5427\n",
"Epoch 601/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 2.2601 - mae: 2.6801 - mape: 1.7091 - r2: 0.7446 - rmse: 5.3516 - val_loss: 1.8207 - val_mae: 2.2099 - val_mape: 1.4252 - val_r2: 0.7959 - val_rmse: 4.3975\n",
"Epoch 602/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 2.2579 - mae: 2.6769 - mape: 1.7077 - r2: 0.7431 - rmse: 5.3323 - val_loss: 1.7587 - val_mae: 2.1529 - val_mape: 1.3870 - val_r2: 0.7998 - val_rmse: 4.3547\n",
"Epoch 603/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 2.2374 - mae: 2.6536 - mape: 1.6861 - r2: 0.7474 - rmse: 5.3085 - val_loss: 1.7986 - val_mae: 2.1841 - val_mape: 1.4143 - val_r2: 0.7946 - val_rmse: 4.4116\n",
"Epoch 604/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 2.2124 - mae: 2.6292 - mape: 1.6793 - r2: 0.7501 - rmse: 5.2582 - val_loss: 1.8189 - val_mae: 2.2132 - val_mape: 1.4302 - val_r2: 0.7975 - val_rmse: 4.3796\n",
"Epoch 605/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 2.1823 - mae: 2.5997 - mape: 1.6500 - r2: 0.7671 - rmse: 5.1210 - val_loss: 1.8211 - val_mae: 2.2184 - val_mape: 1.4342 - val_r2: 0.7954 - val_rmse: 4.4029\n",
"Epoch 606/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 2.2170 - mae: 2.6333 - mape: 1.6771 - r2: 0.7454 - rmse: 5.2684 - val_loss: 1.7918 - val_mae: 2.1747 - val_mape: 1.4056 - val_r2: 0.7966 - val_rmse: 4.3900\n",
"Epoch 607/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 2.1781 - mae: 2.5936 - mape: 1.6533 - r2: 0.7597 - rmse: 5.1675 - val_loss: 1.7967 - val_mae: 2.1901 - val_mape: 1.4102 - val_r2: 0.7872 - val_rmse: 4.4903\n",
"Epoch 608/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 2.2149 - mae: 2.6307 - mape: 1.6764 - r2: 0.7484 - rmse: 5.3097 - val_loss: 1.8532 - val_mae: 2.2443 - val_mape: 1.4583 - val_r2: 0.7787 - val_rmse: 4.5791\n",
"Epoch 609/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 2.2072 - mae: 2.6218 - mape: 1.6755 - r2: 0.7491 - rmse: 5.2804 - val_loss: 1.7286 - val_mae: 2.1123 - val_mape: 1.3620 - val_r2: 0.8040 - val_rmse: 4.3092\n",
"Epoch 610/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 2.1686 - mae: 2.5851 - mape: 1.6425 - r2: 0.7630 - rmse: 5.1470 - val_loss: 1.7119 - val_mae: 2.1038 - val_mape: 1.3565 - val_r2: 0.8095 - val_rmse: 4.2477\n",
"Epoch 611/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 2.1730 - mae: 2.5892 - mape: 1.6508 - r2: 0.7610 - rmse: 5.1730 - val_loss: 1.7918 - val_mae: 2.1814 - val_mape: 1.4081 - val_r2: 0.7919 - val_rmse: 4.4395\n",
"Epoch 612/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 2.1467 - mae: 2.5612 - mape: 1.6328 - r2: 0.7585 - rmse: 5.1687 - val_loss: 1.7880 - val_mae: 2.1828 - val_mape: 1.4132 - val_r2: 0.7940 - val_rmse: 4.4179\n",
"Epoch 613/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 2.1625 - mae: 2.5769 - mape: 1.6405 - r2: 0.7560 - rmse: 5.2159 - val_loss: 1.7316 - val_mae: 2.1202 - val_mape: 1.3724 - val_r2: 0.8001 - val_rmse: 4.3520\n",
"Epoch 614/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 2.1730 - mae: 2.5884 - mape: 1.6540 - r2: 0.7600 - rmse: 5.1921 - val_loss: 1.7652 - val_mae: 2.1560 - val_mape: 1.3989 - val_r2: 0.7917 - val_rmse: 4.4424\n",
"Epoch 615/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step - loss: 2.1538 - mae: 2.5693 - mape: 1.6398 - r2: 0.7562 - rmse: 5.2199 - val_loss: 1.7501 - val_mae: 2.1370 - val_mape: 1.3752 - val_r2: 0.7873 - val_rmse: 4.4892\n",
"Epoch 616/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step - loss: 2.1359 - mae: 2.5475 - mape: 1.6233 - r2: 0.7607 - rmse: 5.1628 - val_loss: 1.7255 - val_mae: 2.1106 - val_mape: 1.3657 - val_r2: 0.7989 - val_rmse: 4.3646\n",
"Epoch 617/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step - loss: 2.1263 - mae: 2.5395 - mape: 1.6183 - r2: 0.7607 - rmse: 5.1821 - val_loss: 1.8260 - val_mae: 2.2272 - val_mape: 1.4472 - val_r2: 0.7904 - val_rmse: 4.4556\n",
"Epoch 618/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 2.1178 - mae: 2.5323 - mape: 1.6155 - r2: 0.7603 - rmse: 5.1345 - val_loss: 1.8586 - val_mae: 2.2562 - val_mape: 1.4688 - val_r2: 0.7902 - val_rmse: 4.4585\n",
"Epoch 619/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 2.1472 - mae: 2.5651 - mape: 1.6416 - r2: 0.7560 - rmse: 5.2109 - val_loss: 1.6979 - val_mae: 2.0835 - val_mape: 1.3453 - val_r2: 0.8080 - val_rmse: 4.2644\n",
"Epoch 620/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 2.1087 - mae: 2.5217 - mape: 1.6080 - r2: 0.7643 - rmse: 5.1416 - val_loss: 1.6593 - val_mae: 2.0510 - val_mape: 1.3210 - val_r2: 0.8184 - val_rmse: 4.1480\n",
"Epoch 621/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 2.0879 - mae: 2.5014 - mape: 1.5916 - r2: 0.7717 - rmse: 5.0566 - val_loss: 1.6568 - val_mae: 2.0379 - val_mape: 1.3193 - val_r2: 0.8109 - val_rmse: 4.2326\n",
"Epoch 622/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 2.0870 - mae: 2.4974 - mape: 1.5918 - r2: 0.7631 - rmse: 5.1221 - val_loss: 1.6541 - val_mae: 2.0382 - val_mape: 1.3193 - val_r2: 0.8131 - val_rmse: 4.2072\n",
"Epoch 623/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 2.0705 - mae: 2.4822 - mape: 1.5859 - r2: 0.7677 - rmse: 5.0919 - val_loss: 1.6990 - val_mae: 2.0828 - val_mape: 1.3490 - val_r2: 0.8094 - val_rmse: 4.2497\n",
"Epoch 624/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 2.0765 - mae: 2.4880 - mape: 1.5860 - r2: 0.7626 - rmse: 5.1689 - val_loss: 1.6514 - val_mae: 2.0334 - val_mape: 1.3133 - val_r2: 0.8134 - val_rmse: 4.2048\n",
"Epoch 625/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 49ms/step - loss: 2.0528 - mae: 2.4625 - mape: 1.5731 - r2: 0.7683 - rmse: 5.0835 - val_loss: 1.6415 - val_mae: 2.0261 - val_mape: 1.3141 - val_r2: 0.8043 - val_rmse: 4.3055\n",
"Epoch 626/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 2.0570 - mae: 2.4680 - mape: 1.5722 - r2: 0.7683 - rmse: 5.0467 - val_loss: 1.6739 - val_mae: 2.0536 - val_mape: 1.3320 - val_r2: 0.8105 - val_rmse: 4.2372\n",
"Epoch 627/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 2.0675 - mae: 2.4781 - mape: 1.5829 - r2: 0.7697 - rmse: 5.0794 - val_loss: 1.7249 - val_mae: 2.1117 - val_mape: 1.3754 - val_r2: 0.8079 - val_rmse: 4.2657\n",
"Epoch 628/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 2.0920 - mae: 2.5047 - mape: 1.6044 - r2: 0.7693 - rmse: 5.0809 - val_loss: 1.6258 - val_mae: 2.0045 - val_mape: 1.2937 - val_r2: 0.8126 - val_rmse: 4.2139\n",
"Epoch 629/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 2.0363 - mae: 2.4466 - mape: 1.5619 - r2: 0.7747 - rmse: 5.0292 - val_loss: 1.6237 - val_mae: 2.0019 - val_mape: 1.2931 - val_r2: 0.7997 - val_rmse: 4.3564\n",
"Epoch 630/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step - loss: 2.0150 - mae: 2.4278 - mape: 1.5472 - r2: 0.7800 - rmse: 4.9627 - val_loss: 1.6333 - val_mae: 2.0095 - val_mape: 1.3036 - val_r2: 0.8174 - val_rmse: 4.1589\n",
"Epoch 631/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 2.0467 - mae: 2.4576 - mape: 1.5704 - r2: 0.7734 - rmse: 5.0248 - val_loss: 1.6818 - val_mae: 2.0649 - val_mape: 1.3390 - val_r2: 0.8091 - val_rmse: 4.2521\n",
"Epoch 632/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 2.0299 - mae: 2.4416 - mape: 1.5610 - r2: 0.7724 - rmse: 5.0379 - val_loss: 1.6467 - val_mae: 2.0237 - val_mape: 1.3138 - val_r2: 0.8126 - val_rmse: 4.2138\n",
"Epoch 633/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 1.9812 - mae: 2.3930 - mape: 1.5254 - r2: 0.7820 - rmse: 4.9389 - val_loss: 1.5378 - val_mae: 1.9157 - val_mape: 1.2368 - val_r2: 0.8217 - val_rmse: 4.1097\n",
"Epoch 634/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 31ms/step - loss: 1.9822 - mae: 2.3897 - mape: 1.5237 - r2: 0.7792 - rmse: 4.9803 - val_loss: 1.5831 - val_mae: 1.9658 - val_mape: 1.2711 - val_r2: 0.8173 - val_rmse: 4.1603\n",
"Epoch 635/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 32ms/step - loss: 2.0007 - mae: 2.4091 - mape: 1.5417 - r2: 0.7764 - rmse: 5.0078 - val_loss: 1.5839 - val_mae: 1.9680 - val_mape: 1.2721 - val_r2: 0.8207 - val_rmse: 4.1216\n",
"Epoch 636/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 32ms/step - loss: 2.0103 - mae: 2.4196 - mape: 1.5418 - r2: 0.7761 - rmse: 4.9991 - val_loss: 1.6290 - val_mae: 2.0106 - val_mape: 1.3021 - val_r2: 0.8151 - val_rmse: 4.1850\n",
"Epoch 637/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 32ms/step - loss: 1.9985 - mae: 2.4084 - mape: 1.5420 - r2: 0.7792 - rmse: 4.9715 - val_loss: 1.6498 - val_mae: 2.0400 - val_mape: 1.3213 - val_r2: 0.8062 - val_rmse: 4.2852\n",
"Epoch 638/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step - loss: 2.0196 - mae: 2.4298 - mape: 1.5543 - r2: 0.7793 - rmse: 4.9664 - val_loss: 1.6622 - val_mae: 2.0484 - val_mape: 1.3295 - val_r2: 0.8156 - val_rmse: 4.1796\n",
"Epoch 639/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 1.9817 - mae: 2.3943 - mape: 1.5264 - r2: 0.7820 - rmse: 4.9383 - val_loss: 1.5041 - val_mae: 1.8772 - val_mape: 1.2097 - val_r2: 0.8222 - val_rmse: 4.1037\n",
"Epoch 640/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 1.9904 - mae: 2.3995 - mape: 1.5356 - r2: 0.7808 - rmse: 4.9411 - val_loss: 1.5412 - val_mae: 1.9275 - val_mape: 1.2389 - val_r2: 0.8234 - val_rmse: 4.0898\n",
"Epoch 641/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step - loss: 1.9504 - mae: 2.3616 - mape: 1.5103 - r2: 0.7807 - rmse: 4.9038 - val_loss: 1.5772 - val_mae: 1.9908 - val_mape: 1.2880 - val_r2: 0.8227 - val_rmse: 4.0983\n",
"Epoch 642/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 2.0064 - mae: 2.4221 - mape: 1.5495 - r2: 0.7802 - rmse: 4.9731 - val_loss: 1.5284 - val_mae: 1.9179 - val_mape: 1.2379 - val_r2: 0.8291 - val_rmse: 4.0237\n",
"Epoch 643/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 1.9393 - mae: 2.3465 - mape: 1.4957 - r2: 0.7870 - rmse: 4.8431 - val_loss: 1.5621 - val_mae: 1.9337 - val_mape: 1.2514 - val_r2: 0.8199 - val_rmse: 4.1301\n",
"Epoch 644/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 1.9158 - mae: 2.3208 - mape: 1.4800 - r2: 0.7877 - rmse: 4.8415 - val_loss: 1.5104 - val_mae: 1.8850 - val_mape: 1.2210 - val_r2: 0.8291 - val_rmse: 4.0239\n",
"Epoch 645/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 1.9190 - mae: 2.3230 - mape: 1.4859 - r2: 0.7839 - rmse: 4.8969 - val_loss: 1.5204 - val_mae: 1.8892 - val_mape: 1.2254 - val_r2: 0.8314 - val_rmse: 3.9960\n",
"Epoch 646/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 1.9641 - mae: 2.3722 - mape: 1.5201 - r2: 0.7782 - rmse: 4.9750 - val_loss: 1.5963 - val_mae: 1.9869 - val_mape: 1.2885 - val_r2: 0.8127 - val_rmse: 4.2122\n",
"Epoch 647/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 1.9429 - mae: 2.3515 - mape: 1.5049 - r2: 0.7833 - rmse: 4.9285 - val_loss: 1.5076 - val_mae: 1.8924 - val_mape: 1.2243 - val_r2: 0.8273 - val_rmse: 4.0449\n",
"Epoch 648/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 1.9028 - mae: 2.3086 - mape: 1.4728 - r2: 0.7874 - rmse: 4.8495 - val_loss: 1.4875 - val_mae: 1.8648 - val_mape: 1.2031 - val_r2: 0.8314 - val_rmse: 3.9968\n",
"Epoch 649/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 1.9210 - mae: 2.3297 - mape: 1.4911 - r2: 0.7857 - rmse: 4.8946 - val_loss: 1.4460 - val_mae: 1.8182 - val_mape: 1.1692 - val_r2: 0.8297 - val_rmse: 4.0166\n",
"Epoch 650/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 1.9060 - mae: 2.3116 - mape: 1.4763 - r2: 0.7863 - rmse: 4.8768 - val_loss: 1.4573 - val_mae: 1.8193 - val_mape: 1.1755 - val_r2: 0.8356 - val_rmse: 3.9463\n",
"Epoch 651/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 1.8849 - mae: 2.2874 - mape: 1.4630 - r2: 0.7841 - rmse: 4.8814 - val_loss: 1.4092 - val_mae: 1.7717 - val_mape: 1.1440 - val_r2: 0.8510 - val_rmse: 3.7565\n",
"Epoch 652/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 1.8908 - mae: 2.2931 - mape: 1.4696 - r2: 0.7846 - rmse: 4.8724 - val_loss: 1.5101 - val_mae: 1.8807 - val_mape: 1.2208 - val_r2: 0.8259 - val_rmse: 4.0616\n",
"Epoch 653/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 31ms/step - loss: 1.8548 - mae: 2.2569 - mape: 1.4402 - r2: 0.7880 - rmse: 4.8431 - val_loss: 1.4822 - val_mae: 1.8469 - val_mape: 1.1967 - val_r2: 0.8251 - val_rmse: 4.0703\n",
"Epoch 654/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 1.8464 - mae: 2.2465 - mape: 1.4327 - r2: 0.7939 - rmse: 4.7898 - val_loss: 1.4300 - val_mae: 1.7903 - val_mape: 1.1504 - val_r2: 0.8372 - val_rmse: 3.9266\n",
"Epoch 655/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 1.8712 - mae: 2.2737 - mape: 1.4567 - r2: 0.7919 - rmse: 4.8329 - val_loss: 1.5442 - val_mae: 1.9202 - val_mape: 1.2424 - val_r2: 0.8186 - val_rmse: 4.1458\n",
"Epoch 656/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 1.8722 - mae: 2.2765 - mape: 1.4539 - r2: 0.7900 - rmse: 4.8380 - val_loss: 1.4455 - val_mae: 1.8127 - val_mape: 1.1728 - val_r2: 0.8319 - val_rmse: 3.9904\n",
"Epoch 657/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 1.8232 - mae: 2.2225 - mape: 1.4215 - r2: 0.7957 - rmse: 4.7828 - val_loss: 1.4803 - val_mae: 1.8464 - val_mape: 1.1919 - val_r2: 0.8207 - val_rmse: 4.1209\n",
"Epoch 658/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 1.8509 - mae: 2.2524 - mape: 1.4413 - r2: 0.7932 - rmse: 4.8239 - val_loss: 1.3595 - val_mae: 1.7171 - val_mape: 1.1002 - val_r2: 0.8393 - val_rmse: 3.9018\n",
"Epoch 659/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 1.8450 - mae: 2.2460 - mape: 1.4356 - r2: 0.7917 - rmse: 4.8187 - val_loss: 1.3588 - val_mae: 1.7229 - val_mape: 1.1098 - val_r2: 0.8439 - val_rmse: 3.8457\n",
"Epoch 660/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 1.8268 - mae: 2.2313 - mape: 1.4232 - r2: 0.8014 - rmse: 4.7205 - val_loss: 1.4582 - val_mae: 1.8691 - val_mape: 1.2076 - val_r2: 0.8530 - val_rmse: 3.7312\n",
"Epoch 661/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 1.8216 - mae: 2.2263 - mape: 1.4244 - r2: 0.7979 - rmse: 4.7510 - val_loss: 1.3919 - val_mae: 1.7715 - val_mape: 1.1428 - val_r2: 0.8411 - val_rmse: 3.8793\n",
"Epoch 662/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 1.8696 - mae: 2.2762 - mape: 1.4580 - r2: 0.7934 - rmse: 4.8136 - val_loss: 1.3956 - val_mae: 1.7654 - val_mape: 1.1383 - val_r2: 0.8461 - val_rmse: 3.8185\n",
"Epoch 663/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 1.8427 - mae: 2.2498 - mape: 1.4407 - r2: 0.7962 - rmse: 4.7417 - val_loss: 1.3749 - val_mae: 1.7362 - val_mape: 1.1183 - val_r2: 0.8445 - val_rmse: 3.8379\n",
"Epoch 664/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 1.7951 - mae: 2.1956 - mape: 1.4012 - r2: 0.8066 - rmse: 4.6593 - val_loss: 1.5034 - val_mae: 1.8914 - val_mape: 1.2327 - val_r2: 0.8313 - val_rmse: 3.9978\n",
"Epoch 665/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 1.8084 - mae: 2.2117 - mape: 1.4151 - r2: 0.8005 - rmse: 4.7196 - val_loss: 1.4712 - val_mae: 1.8322 - val_mape: 1.1920 - val_r2: 0.8178 - val_rmse: 4.1549\n",
"Epoch 666/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 1.8087 - mae: 2.2086 - mape: 1.4171 - r2: 0.7982 - rmse: 4.7598 - val_loss: 1.3582 - val_mae: 1.7157 - val_mape: 1.1064 - val_r2: 0.8397 - val_rmse: 3.8971\n",
"Epoch 667/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 1.7904 - mae: 2.1877 - mape: 1.4023 - r2: 0.7961 - rmse: 4.7806 - val_loss: 1.3846 - val_mae: 1.7421 - val_mape: 1.1186 - val_r2: 0.8366 - val_rmse: 3.9338\n",
"Epoch 668/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 1.7366 - mae: 2.1347 - mape: 1.3636 - r2: 0.8078 - rmse: 4.6112 - val_loss: 1.3999 - val_mae: 1.7492 - val_mape: 1.1336 - val_r2: 0.8293 - val_rmse: 4.0212\n",
"Epoch 669/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 1.7761 - mae: 2.1718 - mape: 1.3919 - r2: 0.7988 - rmse: 4.7423 - val_loss: 1.3894 - val_mae: 1.7436 - val_mape: 1.1232 - val_r2: 0.8331 - val_rmse: 3.9760\n",
"Epoch 670/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step - loss: 1.7711 - mae: 2.1664 - mape: 1.3844 - r2: 0.7935 - rmse: 4.7446 - val_loss: 1.4171 - val_mae: 1.7692 - val_mape: 1.1450 - val_r2: 0.8290 - val_rmse: 4.0250\n",
"Epoch 671/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step - loss: 1.7480 - mae: 2.1463 - mape: 1.3735 - r2: 0.8041 - rmse: 4.6703 - val_loss: 1.4815 - val_mae: 1.8655 - val_mape: 1.2082 - val_r2: 0.8346 - val_rmse: 3.9583\n",
"Epoch 672/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 1.7527 - mae: 2.1492 - mape: 1.3722 - r2: 0.8062 - rmse: 4.6515 - val_loss: 1.3935 - val_mae: 1.7635 - val_mape: 1.1417 - val_r2: 0.8426 - val_rmse: 3.8614\n",
"Epoch 673/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 1.7178 - mae: 2.1121 - mape: 1.3463 - r2: 0.8109 - rmse: 4.5931 - val_loss: 1.2985 - val_mae: 1.6483 - val_mape: 1.0649 - val_r2: 0.8516 - val_rmse: 3.7499\n",
"Epoch 674/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 1.7380 - mae: 2.1313 - mape: 1.3645 - r2: 0.8047 - rmse: 4.6473 - val_loss: 1.3617 - val_mae: 1.7136 - val_mape: 1.1104 - val_r2: 0.8433 - val_rmse: 3.8529\n",
"Epoch 675/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 1.7386 - mae: 2.1345 - mape: 1.3644 - r2: 0.8049 - rmse: 4.6536 - val_loss: 1.3621 - val_mae: 1.7115 - val_mape: 1.1059 - val_r2: 0.8340 - val_rmse: 3.9657\n",
"Epoch 676/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 1.7202 - mae: 2.1113 - mape: 1.3520 - r2: 0.8045 - rmse: 4.6804 - val_loss: 1.2746 - val_mae: 1.6264 - val_mape: 1.0451 - val_r2: 0.8562 - val_rmse: 3.6911\n",
"Epoch 677/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 1.7359 - mae: 2.1305 - mape: 1.3621 - r2: 0.8065 - rmse: 4.6179 - val_loss: 1.3608 - val_mae: 1.7064 - val_mape: 1.1082 - val_r2: 0.8382 - val_rmse: 3.9151\n",
"Epoch 678/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 1.7258 - mae: 2.1188 - mape: 1.3559 - r2: 0.8092 - rmse: 4.6291 - val_loss: 1.3521 - val_mae: 1.7032 - val_mape: 1.1020 - val_r2: 0.8323 - val_rmse: 3.9863\n",
"Epoch 679/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 1.6677 - mae: 2.0596 - mape: 1.3134 - r2: 0.8158 - rmse: 4.5263 - val_loss: 1.2880 - val_mae: 1.6373 - val_mape: 1.0536 - val_r2: 0.8388 - val_rmse: 3.9081\n",
"Epoch 680/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 1.7181 - mae: 2.1124 - mape: 1.3502 - r2: 0.8089 - rmse: 4.5933 - val_loss: 1.3338 - val_mae: 1.6794 - val_mape: 1.0877 - val_r2: 0.8422 - val_rmse: 3.8660\n",
"Epoch 681/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 1.7045 - mae: 2.0969 - mape: 1.3455 - r2: 0.8091 - rmse: 4.6310 - val_loss: 1.2958 - val_mae: 1.6337 - val_mape: 1.0577 - val_r2: 0.8402 - val_rmse: 3.8904\n",
"Epoch 682/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step - loss: 1.6834 - mae: 2.0761 - mape: 1.3256 - r2: 0.8135 - rmse: 4.5704 - val_loss: 1.2768 - val_mae: 1.6200 - val_mape: 1.0417 - val_r2: 0.8482 - val_rmse: 3.7917\n",
"Epoch 683/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 32ms/step - loss: 1.7042 - mae: 2.0986 - mape: 1.3453 - r2: 0.8128 - rmse: 4.5695 - val_loss: 1.3672 - val_mae: 1.7448 - val_mape: 1.1305 - val_r2: 0.8404 - val_rmse: 3.8887\n",
"Epoch 684/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 1.7172 - mae: 2.1131 - mape: 1.3520 - r2: 0.8119 - rmse: 4.6006 - val_loss: 1.2371 - val_mae: 1.5815 - val_mape: 1.0145 - val_r2: 0.8542 - val_rmse: 3.7167\n",
"Epoch 685/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 1.6939 - mae: 2.0867 - mape: 1.3350 - r2: 0.8098 - rmse: 4.6198 - val_loss: 1.2956 - val_mae: 1.6452 - val_mape: 1.0550 - val_r2: 0.8447 - val_rmse: 3.8359\n",
"Epoch 686/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 30ms/step - loss: 1.6797 - mae: 2.0724 - mape: 1.3300 - r2: 0.8110 - rmse: 4.5901 - val_loss: 1.2770 - val_mae: 1.6246 - val_mape: 1.0496 - val_r2: 0.8570 - val_rmse: 3.6801\n",
"Epoch 687/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 1.6803 - mae: 2.0707 - mape: 1.3282 - r2: 0.8046 - rmse: 4.6799 - val_loss: 1.2254 - val_mae: 1.5609 - val_mape: 1.0025 - val_r2: 0.8464 - val_rmse: 3.8146\n",
"Epoch 688/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 1.6785 - mae: 2.0704 - mape: 1.3213 - r2: 0.8128 - rmse: 4.5749 - val_loss: 1.2736 - val_mae: 1.6290 - val_mape: 1.0494 - val_r2: 0.8566 - val_rmse: 3.6862\n",
"Epoch 689/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 1.6680 - mae: 2.0587 - mape: 1.3240 - r2: 0.8156 - rmse: 4.5355 - val_loss: 1.2325 - val_mae: 1.5731 - val_mape: 1.0163 - val_r2: 0.8536 - val_rmse: 3.7246\n",
"Epoch 690/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step - loss: 1.6326 - mae: 2.0217 - mape: 1.2891 - r2: 0.8173 - rmse: 4.5054 - val_loss: 1.2924 - val_mae: 1.6771 - val_mape: 1.0846 - val_r2: 0.8552 - val_rmse: 3.7042\n",
"Epoch 691/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 1.6575 - mae: 2.0520 - mape: 1.3124 - r2: 0.8160 - rmse: 4.5395 - val_loss: 1.2247 - val_mae: 1.5587 - val_mape: 1.0039 - val_r2: 0.8469 - val_rmse: 3.8078\n",
"Epoch 692/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step - loss: 1.6538 - mae: 2.0449 - mape: 1.3151 - r2: 0.8119 - rmse: 4.5893 - val_loss: 1.2594 - val_mae: 1.6252 - val_mape: 1.0458 - val_r2: 0.8655 - val_rmse: 3.5693\n",
"Epoch 693/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step - loss: 1.6442 - mae: 2.0352 - mape: 1.2999 - r2: 0.8150 - rmse: 4.5405 - val_loss: 1.1748 - val_mae: 1.5072 - val_mape: 0.9656 - val_r2: 0.8545 - val_rmse: 3.7123\n",
"Epoch 694/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 1.6972 - mae: 2.0864 - mape: 1.3393 - r2: 0.8013 - rmse: 4.7089 - val_loss: 1.2998 - val_mae: 1.6679 - val_mape: 1.0837 - val_r2: 0.8564 - val_rmse: 3.6878\n",
"Epoch 695/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 1.6053 - mae: 1.9926 - mape: 1.2758 - r2: 0.8233 - rmse: 4.4362 - val_loss: 1.2256 - val_mae: 1.5578 - val_mape: 1.0105 - val_r2: 0.8547 - val_rmse: 3.7098\n",
"Epoch 696/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 1.6298 - mae: 2.0180 - mape: 1.2922 - r2: 0.8171 - rmse: 4.5343 - val_loss: 1.2783 - val_mae: 1.6169 - val_mape: 1.0430 - val_r2: 0.8475 - val_rmse: 3.8011\n",
"Epoch 697/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 1.6033 - mae: 1.9886 - mape: 1.2734 - r2: 0.8148 - rmse: 4.5299 - val_loss: 1.2711 - val_mae: 1.6122 - val_mape: 1.0469 - val_r2: 0.8445 - val_rmse: 3.8385\n",
"Epoch 698/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step - loss: 1.6163 - mae: 2.0021 - mape: 1.2841 - r2: 0.8171 - rmse: 4.5028 - val_loss: 1.2374 - val_mae: 1.5745 - val_mape: 1.0159 - val_r2: 0.8410 - val_rmse: 3.8810\n",
"Epoch 699/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 1.6319 - mae: 2.0183 - mape: 1.2942 - r2: 0.8156 - rmse: 4.5311 - val_loss: 1.1830 - val_mae: 1.5094 - val_mape: 0.9730 - val_r2: 0.8565 - val_rmse: 3.6870\n",
"Epoch 700/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 1.6215 - mae: 2.0095 - mape: 1.2879 - r2: 0.8157 - rmse: 4.5467 - val_loss: 1.2480 - val_mae: 1.6140 - val_mape: 1.0431 - val_r2: 0.8525 - val_rmse: 3.7377\n",
"Epoch 701/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 1.6195 - mae: 2.0093 - mape: 1.2900 - r2: 0.8186 - rmse: 4.5151 - val_loss: 1.1960 - val_mae: 1.5319 - val_mape: 0.9863 - val_r2: 0.8585 - val_rmse: 3.6606\n",
"Epoch 702/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step - loss: 1.5807 - mae: 1.9705 - mape: 1.2629 - r2: 0.8244 - rmse: 4.4320 - val_loss: 1.2556 - val_mae: 1.6050 - val_mape: 1.0354 - val_r2: 0.8418 - val_rmse: 3.8712\n",
"Epoch 703/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 1.6037 - mae: 1.9897 - mape: 1.2792 - r2: 0.8206 - rmse: 4.4809 - val_loss: 1.1558 - val_mae: 1.4876 - val_mape: 0.9507 - val_r2: 0.8675 - val_rmse: 3.5430\n",
"Epoch 704/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step - loss: 1.6727 - mae: 2.0641 - mape: 1.3262 - r2: 0.8080 - rmse: 4.5947 - val_loss: 1.2592 - val_mae: 1.6147 - val_mape: 1.0471 - val_r2: 0.8461 - val_rmse: 3.8179\n",
"Epoch 705/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 1.6241 - mae: 2.0169 - mape: 1.2936 - r2: 0.8178 - rmse: 4.4840 - val_loss: 1.2587 - val_mae: 1.6211 - val_mape: 1.0484 - val_r2: 0.8531 - val_rmse: 3.7300\n",
"Epoch 706/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 1.6130 - mae: 2.0045 - mape: 1.2892 - r2: 0.8204 - rmse: 4.4712 - val_loss: 1.2542 - val_mae: 1.5967 - val_mape: 1.0313 - val_r2: 0.8412 - val_rmse: 3.8783\n",
"Epoch 707/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 1.6090 - mae: 1.9956 - mape: 1.2798 - r2: 0.8110 - rmse: 4.5299 - val_loss: 1.2942 - val_mae: 1.6402 - val_mape: 1.0668 - val_r2: 0.8376 - val_rmse: 3.9223\n",
"Epoch 708/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 1.5755 - mae: 1.9605 - mape: 1.2556 - r2: 0.8241 - rmse: 4.4400 - val_loss: 1.1515 - val_mae: 1.4820 - val_mape: 0.9548 - val_r2: 0.8653 - val_rmse: 3.5719\n",
"Epoch 709/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step - loss: 1.5938 - mae: 1.9789 - mape: 1.2674 - r2: 0.8219 - rmse: 4.4724 - val_loss: 1.1831 - val_mae: 1.5225 - val_mape: 0.9848 - val_r2: 0.8596 - val_rmse: 3.6469\n",
"Epoch 710/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 32ms/step - loss: 1.5737 - mae: 1.9591 - mape: 1.2563 - r2: 0.8249 - rmse: 4.4280 - val_loss: 1.1935 - val_mae: 1.5491 - val_mape: 1.0039 - val_r2: 0.8571 - val_rmse: 3.6797\n",
"Epoch 711/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step - loss: 1.6074 - mae: 1.9920 - mape: 1.2795 - r2: 0.8174 - rmse: 4.4899 - val_loss: 1.2234 - val_mae: 1.5503 - val_mape: 1.0011 - val_r2: 0.8442 - val_rmse: 3.8416\n",
"Epoch 712/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step - loss: 1.5629 - mae: 1.9445 - mape: 1.2484 - r2: 0.8229 - rmse: 4.4407 - val_loss: 1.1414 - val_mae: 1.4690 - val_mape: 0.9416 - val_r2: 0.8579 - val_rmse: 3.6696\n",
"Epoch 713/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 1.5483 - mae: 1.9306 - mape: 1.2387 - r2: 0.8236 - rmse: 4.4430 - val_loss: 1.2307 - val_mae: 1.5618 - val_mape: 1.0017 - val_r2: 0.8376 - val_rmse: 3.9222\n",
"Epoch 714/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 1.5639 - mae: 1.9455 - mape: 1.2495 - r2: 0.8224 - rmse: 4.4649 - val_loss: 1.2287 - val_mae: 1.5578 - val_mape: 1.0028 - val_r2: 0.8420 - val_rmse: 3.8688\n",
"Epoch 715/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 1.5645 - mae: 1.9476 - mape: 1.2493 - r2: 0.8271 - rmse: 4.3944 - val_loss: 1.1531 - val_mae: 1.4910 - val_mape: 0.9583 - val_r2: 0.8735 - val_rmse: 3.4622\n",
"Epoch 716/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step - loss: 1.5506 - mae: 1.9350 - mape: 1.2416 - r2: 0.8264 - rmse: 4.4202 - val_loss: 1.1733 - val_mae: 1.5101 - val_mape: 0.9672 - val_r2: 0.8579 - val_rmse: 3.6691\n",
"Epoch 717/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step - loss: 1.5360 - mae: 1.9161 - mape: 1.2303 - r2: 0.8238 - rmse: 4.4192 - val_loss: 1.1429 - val_mae: 1.4698 - val_mape: 0.9407 - val_r2: 0.8504 - val_rmse: 3.7649\n",
"Epoch 718/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 1.5251 - mae: 1.9072 - mape: 1.2253 - r2: 0.8284 - rmse: 4.3921 - val_loss: 1.2353 - val_mae: 1.5677 - val_mape: 1.0123 - val_r2: 0.8576 - val_rmse: 3.6726\n",
"Epoch 719/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 1.5248 - mae: 1.9039 - mape: 1.2186 - r2: 0.8270 - rmse: 4.4013 - val_loss: 1.1640 - val_mae: 1.4898 - val_mape: 0.9582 - val_r2: 0.8644 - val_rmse: 3.5836\n",
"Epoch 720/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 1.5530 - mae: 1.9334 - mape: 1.2416 - r2: 0.8228 - rmse: 4.4378 - val_loss: 1.1722 - val_mae: 1.4970 - val_mape: 0.9684 - val_r2: 0.8627 - val_rmse: 3.6058\n",
"Epoch 721/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 1.5236 - mae: 1.9046 - mape: 1.2260 - r2: 0.8221 - rmse: 4.4296 - val_loss: 1.2081 - val_mae: 1.5434 - val_mape: 0.9948 - val_r2: 0.8532 - val_rmse: 3.7288\n",
"Epoch 722/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 1.5707 - mae: 1.9531 - mape: 1.2570 - r2: 0.8203 - rmse: 4.4674 - val_loss: 1.2265 - val_mae: 1.5624 - val_mape: 1.0110 - val_r2: 0.8559 - val_rmse: 3.6946\n",
"Epoch 723/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 1.5558 - mae: 1.9378 - mape: 1.2440 - r2: 0.8236 - rmse: 4.4423 - val_loss: 1.1532 - val_mae: 1.4806 - val_mape: 0.9596 - val_r2: 0.8628 - val_rmse: 3.6053\n",
"Epoch 724/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 1.5134 - mae: 1.8954 - mape: 1.2172 - r2: 0.8344 - rmse: 4.3113 - val_loss: 1.1452 - val_mae: 1.4722 - val_mape: 0.9427 - val_r2: 0.8702 - val_rmse: 3.5071\n",
"Epoch 725/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 1.4976 - mae: 1.8786 - mape: 1.2013 - r2: 0.8353 - rmse: 4.2597 - val_loss: 1.1370 - val_mae: 1.4653 - val_mape: 0.9440 - val_r2: 0.8639 - val_rmse: 3.5909\n",
"Epoch 726/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step - loss: 1.5523 - mae: 1.9316 - mape: 1.2423 - r2: 0.8200 - rmse: 4.4897 - val_loss: 1.1652 - val_mae: 1.4904 - val_mape: 0.9632 - val_r2: 0.8518 - val_rmse: 3.7466\n",
"Epoch 727/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step - loss: 1.5409 - mae: 1.9206 - mape: 1.2374 - r2: 0.8213 - rmse: 4.4685 - val_loss: 1.1989 - val_mae: 1.5309 - val_mape: 0.9941 - val_r2: 0.8438 - val_rmse: 3.8461\n",
"Epoch 728/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 1.5137 - mae: 1.8922 - mape: 1.2162 - r2: 0.8255 - rmse: 4.3903 - val_loss: 1.1360 - val_mae: 1.4560 - val_mape: 0.9380 - val_r2: 0.8644 - val_rmse: 3.5838\n",
"Epoch 729/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 1.5136 - mae: 1.8924 - mape: 1.2155 - r2: 0.8298 - rmse: 4.3583 - val_loss: 1.2177 - val_mae: 1.5579 - val_mape: 1.0097 - val_r2: 0.8536 - val_rmse: 3.7235\n",
"Epoch 730/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 1.5039 - mae: 1.8817 - mape: 1.2062 - r2: 0.8323 - rmse: 4.3390 - val_loss: 1.2005 - val_mae: 1.5656 - val_mape: 1.0205 - val_r2: 0.8641 - val_rmse: 3.5875\n",
"Epoch 731/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 1.5096 - mae: 1.8908 - mape: 1.2102 - r2: 0.8338 - rmse: 4.3193 - val_loss: 1.0764 - val_mae: 1.4127 - val_mape: 0.9129 - val_r2: 0.8880 - val_rmse: 3.2573\n",
"Epoch 732/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step - loss: 1.5391 - mae: 1.9212 - mape: 1.2373 - r2: 0.8280 - rmse: 4.3590 - val_loss: 1.1502 - val_mae: 1.4723 - val_mape: 0.9427 - val_r2: 0.8594 - val_rmse: 3.6493\n",
"Epoch 733/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 1.5010 - mae: 1.8803 - mape: 1.2082 - r2: 0.8287 - rmse: 4.3492 - val_loss: 1.1619 - val_mae: 1.4917 - val_mape: 0.9610 - val_r2: 0.8522 - val_rmse: 3.7422\n",
"Epoch 734/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 1.5033 - mae: 1.8804 - mape: 1.2082 - r2: 0.8315 - rmse: 4.3341 - val_loss: 1.1764 - val_mae: 1.5136 - val_mape: 0.9783 - val_r2: 0.8464 - val_rmse: 3.8139\n",
"Epoch 735/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 1.4963 - mae: 1.8746 - mape: 1.2033 - r2: 0.8312 - rmse: 4.3590 - val_loss: 1.1569 - val_mae: 1.4895 - val_mape: 0.9703 - val_r2: 0.8623 - val_rmse: 3.6115\n",
"Epoch 736/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 41ms/step - loss: 1.4897 - mae: 1.8677 - mape: 1.2002 - r2: 0.8367 - rmse: 4.2821 - val_loss: 1.2253 - val_mae: 1.5681 - val_mape: 1.0265 - val_r2: 0.8494 - val_rmse: 3.7771\n",
"Epoch 737/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step - loss: 1.4981 - mae: 1.8766 - mape: 1.2064 - r2: 0.8291 - rmse: 4.3590 - val_loss: 1.1157 - val_mae: 1.4454 - val_mape: 0.9362 - val_r2: 0.8605 - val_rmse: 3.6358\n",
"Epoch 738/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 1.4790 - mae: 1.8558 - mape: 1.1905 - r2: 0.8333 - rmse: 4.3012 - val_loss: 1.1510 - val_mae: 1.4846 - val_mape: 0.9541 - val_r2: 0.8746 - val_rmse: 3.4466\n",
"Epoch 739/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 1.4828 - mae: 1.8619 - mape: 1.1961 - r2: 0.8381 - rmse: 4.2447 - val_loss: 1.1650 - val_mae: 1.4899 - val_mape: 0.9670 - val_r2: 0.8558 - val_rmse: 3.6964\n",
"Epoch 740/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 1.4764 - mae: 1.8548 - mape: 1.1931 - r2: 0.8350 - rmse: 4.2666 - val_loss: 1.1798 - val_mae: 1.5357 - val_mape: 0.9965 - val_r2: 0.8710 - val_rmse: 3.4954\n",
"Epoch 741/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 1.4936 - mae: 1.8714 - mape: 1.2013 - r2: 0.8289 - rmse: 4.3563 - val_loss: 1.0982 - val_mae: 1.4232 - val_mape: 0.9166 - val_r2: 0.8664 - val_rmse: 3.5578\n",
"Epoch 742/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 1.5178 - mae: 1.8949 - mape: 1.2246 - r2: 0.8248 - rmse: 4.4158 - val_loss: 1.1888 - val_mae: 1.5159 - val_mape: 0.9897 - val_r2: 0.8606 - val_rmse: 3.6336\n",
"Epoch 743/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 31ms/step - loss: 1.4629 - mae: 1.8403 - mape: 1.1820 - r2: 0.8354 - rmse: 4.2545 - val_loss: 1.1396 - val_mae: 1.4628 - val_mape: 0.9403 - val_r2: 0.8581 - val_rmse: 3.6664\n",
"Epoch 744/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 1.4554 - mae: 1.8305 - mape: 1.1727 - r2: 0.8380 - rmse: 4.2628 - val_loss: 1.1477 - val_mae: 1.4668 - val_mape: 0.9483 - val_r2: 0.8627 - val_rmse: 3.6068\n",
"Epoch 745/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 43ms/step - loss: 1.5033 - mae: 1.8762 - mape: 1.2107 - r2: 0.8218 - rmse: 4.4464 - val_loss: 1.0471 - val_mae: 1.3656 - val_mape: 0.8826 - val_r2: 0.8738 - val_rmse: 3.4576\n",
"Epoch 746/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 1.4871 - mae: 1.8639 - mape: 1.2007 - r2: 0.8320 - rmse: 4.3439 - val_loss: 1.0717 - val_mae: 1.4058 - val_mape: 0.9039 - val_r2: 0.8794 - val_rmse: 3.3804\n",
"Epoch 747/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 1.4576 - mae: 1.8320 - mape: 1.1743 - r2: 0.8386 - rmse: 4.2567 - val_loss: 1.1129 - val_mae: 1.4266 - val_mape: 0.9205 - val_r2: 0.8674 - val_rmse: 3.5446\n",
"Epoch 748/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 1.4479 - mae: 1.8205 - mape: 1.1710 - r2: 0.8356 - rmse: 4.2605 - val_loss: 1.1269 - val_mae: 1.4597 - val_mape: 0.9452 - val_r2: 0.8662 - val_rmse: 3.5599\n",
"Epoch 749/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 1.4768 - mae: 1.8479 - mape: 1.1900 - r2: 0.8332 - rmse: 4.3074 - val_loss: 1.0963 - val_mae: 1.4189 - val_mape: 0.9187 - val_r2: 0.8652 - val_rmse: 3.5732\n",
"Epoch 750/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 1.4175 - mae: 1.7898 - mape: 1.1519 - r2: 0.8376 - rmse: 4.2521 - val_loss: 1.0443 - val_mae: 1.3663 - val_mape: 0.8858 - val_r2: 0.8747 - val_rmse: 3.4452\n",
"Epoch 751/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 1.4677 - mae: 1.8432 - mape: 1.1848 - r2: 0.8310 - rmse: 4.3496 - val_loss: 1.1868 - val_mae: 1.5129 - val_mape: 0.9792 - val_r2: 0.8528 - val_rmse: 3.7337\n",
"Epoch 752/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 1.4862 - mae: 1.8590 - mape: 1.1968 - r2: 0.8329 - rmse: 4.3322 - val_loss: 1.1501 - val_mae: 1.4765 - val_mape: 0.9478 - val_r2: 0.8574 - val_rmse: 3.6754\n",
"Epoch 753/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 1.4546 - mae: 1.8266 - mape: 1.1767 - r2: 0.8326 - rmse: 4.3203 - val_loss: 1.0828 - val_mae: 1.4010 - val_mape: 0.9049 - val_r2: 0.8750 - val_rmse: 3.4416\n",
"Epoch 754/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 1.4113 - mae: 1.7829 - mape: 1.1465 - r2: 0.8423 - rmse: 4.2026 - val_loss: 1.0409 - val_mae: 1.3697 - val_mape: 0.8821 - val_r2: 0.8838 - val_rmse: 3.3171\n",
"Epoch 755/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 1.4419 - mae: 1.8145 - mape: 1.1660 - r2: 0.8391 - rmse: 4.2101 - val_loss: 1.1229 - val_mae: 1.4548 - val_mape: 0.9392 - val_r2: 0.8504 - val_rmse: 3.7641\n",
"Epoch 756/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 1.4571 - mae: 1.8316 - mape: 1.1804 - r2: 0.8331 - rmse: 4.3067 - val_loss: 1.2194 - val_mae: 1.5712 - val_mape: 1.0231 - val_r2: 0.8582 - val_rmse: 3.6647\n",
"Epoch 757/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 1.4687 - mae: 1.8446 - mape: 1.1880 - r2: 0.8336 - rmse: 4.3257 - val_loss: 1.1484 - val_mae: 1.4647 - val_mape: 0.9429 - val_r2: 0.8442 - val_rmse: 3.8415\n",
"Epoch 758/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 1.4142 - mae: 1.7877 - mape: 1.1426 - r2: 0.8427 - rmse: 4.2021 - val_loss: 1.1535 - val_mae: 1.4721 - val_mape: 0.9495 - val_r2: 0.8557 - val_rmse: 3.6977\n",
"Epoch 759/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 1.3912 - mae: 1.7621 - mape: 1.1306 - r2: 0.8456 - rmse: 4.1391 - val_loss: 1.0920 - val_mae: 1.4241 - val_mape: 0.9169 - val_r2: 0.8728 - val_rmse: 3.4710\n",
"Epoch 760/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 1.4366 - mae: 1.8091 - mape: 1.1631 - r2: 0.8372 - rmse: 4.2635 - val_loss: 1.1395 - val_mae: 1.4602 - val_mape: 0.9495 - val_r2: 0.8636 - val_rmse: 3.5942\n",
"Epoch 761/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 1.4205 - mae: 1.7922 - mape: 1.1507 - r2: 0.8401 - rmse: 4.2363 - val_loss: 1.1216 - val_mae: 1.4388 - val_mape: 0.9421 - val_r2: 0.8627 - val_rmse: 3.6063\n",
"Epoch 762/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 33ms/step - loss: 1.4336 - mae: 1.8046 - mape: 1.1633 - r2: 0.8384 - rmse: 4.2421 - val_loss: 1.1659 - val_mae: 1.4999 - val_mape: 0.9736 - val_r2: 0.8599 - val_rmse: 3.6427\n",
"Epoch 763/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 40ms/step - loss: 1.4259 - mae: 1.7972 - mape: 1.1600 - r2: 0.8416 - rmse: 4.2221 - val_loss: 1.1271 - val_mae: 1.4558 - val_mape: 0.9368 - val_r2: 0.8507 - val_rmse: 3.7602\n",
"Epoch 764/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 1.4486 - mae: 1.8203 - mape: 1.1757 - r2: 0.8317 - rmse: 4.3486 - val_loss: 1.1030 - val_mae: 1.4130 - val_mape: 0.9195 - val_r2: 0.8538 - val_rmse: 3.7219\n",
"Epoch 765/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 1.4041 - mae: 1.7739 - mape: 1.1383 - r2: 0.8458 - rmse: 4.1671 - val_loss: 1.0852 - val_mae: 1.4004 - val_mape: 0.9085 - val_r2: 0.8696 - val_rmse: 3.5153\n",
"Epoch 766/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step - loss: 1.4576 - mae: 1.8295 - mape: 1.1785 - r2: 0.8343 - rmse: 4.3052 - val_loss: 1.1168 - val_mae: 1.4492 - val_mape: 0.9413 - val_r2: 0.8639 - val_rmse: 3.5908\n",
"Epoch 767/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 1.3798 - mae: 1.7480 - mape: 1.1225 - r2: 0.8450 - rmse: 4.1595 - val_loss: 1.1307 - val_mae: 1.4732 - val_mape: 0.9578 - val_r2: 0.8610 - val_rmse: 3.6288\n",
"Epoch 768/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 1.4267 - mae: 1.7955 - mape: 1.1532 - r2: 0.8391 - rmse: 4.2400 - val_loss: 1.1821 - val_mae: 1.5080 - val_mape: 0.9836 - val_r2: 0.8473 - val_rmse: 3.8029\n",
"Epoch 769/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 1.4204 - mae: 1.7905 - mape: 1.1507 - r2: 0.8413 - rmse: 4.2066 - val_loss: 1.0413 - val_mae: 1.3503 - val_mape: 0.8780 - val_r2: 0.8717 - val_rmse: 3.4859\n",
"Epoch 770/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 1.3612 - mae: 1.7298 - mape: 1.1087 - r2: 0.8471 - rmse: 4.1471 - val_loss: 1.1202 - val_mae: 1.4378 - val_mape: 0.9340 - val_r2: 0.8605 - val_rmse: 3.6357\n",
"Epoch 771/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 1.4115 - mae: 1.7816 - mape: 1.1481 - r2: 0.8399 - rmse: 4.2203 - val_loss: 1.1239 - val_mae: 1.4369 - val_mape: 0.9334 - val_r2: 0.8569 - val_rmse: 3.6819\n",
"Epoch 772/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 1.3727 - mae: 1.7410 - mape: 1.1189 - r2: 0.8468 - rmse: 4.1431 - val_loss: 1.0894 - val_mae: 1.4218 - val_mape: 0.9277 - val_r2: 0.8768 - val_rmse: 3.4161\n",
"Epoch 773/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 1.4326 - mae: 1.8017 - mape: 1.1635 - r2: 0.8371 - rmse: 4.2743 - val_loss: 1.0800 - val_mae: 1.3943 - val_mape: 0.9069 - val_r2: 0.8672 - val_rmse: 3.5462\n",
"Epoch 774/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 1.3818 - mae: 1.7503 - mape: 1.1259 - r2: 0.8455 - rmse: 4.1593 - val_loss: 1.0401 - val_mae: 1.3605 - val_mape: 0.8804 - val_r2: 0.8785 - val_rmse: 3.3931\n",
"Epoch 775/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 1.3995 - mae: 1.7662 - mape: 1.1355 - r2: 0.8464 - rmse: 4.1506 - val_loss: 1.0587 - val_mae: 1.3765 - val_mape: 0.8931 - val_r2: 0.8844 - val_rmse: 3.3095\n",
"Epoch 776/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 1.3882 - mae: 1.7551 - mape: 1.1303 - r2: 0.8473 - rmse: 4.1444 - val_loss: 1.1169 - val_mae: 1.4585 - val_mape: 0.9418 - val_r2: 0.8575 - val_rmse: 3.6747\n",
"Epoch 777/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 1.3843 - mae: 1.7539 - mape: 1.1310 - r2: 0.8439 - rmse: 4.1655 - val_loss: 1.0745 - val_mae: 1.3887 - val_mape: 0.8936 - val_r2: 0.8619 - val_rmse: 3.6171\n",
"Epoch 778/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 1.3766 - mae: 1.7427 - mape: 1.1206 - r2: 0.8465 - rmse: 4.1412 - val_loss: 1.0887 - val_mae: 1.4129 - val_mape: 0.9243 - val_r2: 0.8695 - val_rmse: 3.5156\n",
"Epoch 779/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 1.3920 - mae: 1.7599 - mape: 1.1362 - r2: 0.8420 - rmse: 4.1899 - val_loss: 1.0268 - val_mae: 1.3646 - val_mape: 0.8845 - val_r2: 0.8950 - val_rmse: 3.1540\n",
"Epoch 780/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 1.3876 - mae: 1.7548 - mape: 1.1306 - r2: 0.8400 - rmse: 4.1942 - val_loss: 1.0981 - val_mae: 1.4304 - val_mape: 0.9240 - val_r2: 0.8682 - val_rmse: 3.5335\n",
"Epoch 781/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 1.3762 - mae: 1.7450 - mape: 1.1249 - r2: 0.8448 - rmse: 4.1555 - val_loss: 1.1212 - val_mae: 1.4456 - val_mape: 0.9421 - val_r2: 0.8645 - val_rmse: 3.5833\n",
"Epoch 782/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 1.3770 - mae: 1.7424 - mape: 1.1242 - r2: 0.8408 - rmse: 4.2068 - val_loss: 1.0812 - val_mae: 1.3916 - val_mape: 0.8997 - val_r2: 0.8702 - val_rmse: 3.5071\n",
"Epoch 783/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 1.4088 - mae: 1.7786 - mape: 1.1478 - r2: 0.8412 - rmse: 4.2209 - val_loss: 1.0465 - val_mae: 1.3642 - val_mape: 0.8835 - val_r2: 0.8748 - val_rmse: 3.4434\n",
"Epoch 784/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step - loss: 1.4178 - mae: 1.7875 - mape: 1.1567 - r2: 0.8353 - rmse: 4.2542 - val_loss: 0.9918 - val_mae: 1.3090 - val_mape: 0.8491 - val_r2: 0.8767 - val_rmse: 3.4183\n",
"Epoch 785/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 1.3691 - mae: 1.7354 - mape: 1.1150 - r2: 0.8453 - rmse: 4.1237 - val_loss: 1.1208 - val_mae: 1.4445 - val_mape: 0.9388 - val_r2: 0.8595 - val_rmse: 3.6488\n",
"Epoch 786/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 1.3957 - mae: 1.7608 - mape: 1.1360 - r2: 0.8415 - rmse: 4.2109 - val_loss: 1.0842 - val_mae: 1.4086 - val_mape: 0.9111 - val_r2: 0.8582 - val_rmse: 3.6646\n",
"Epoch 787/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 1.3622 - mae: 1.7286 - mape: 1.1106 - r2: 0.8478 - rmse: 4.1237 - val_loss: 1.0667 - val_mae: 1.3717 - val_mape: 0.8829 - val_r2: 0.8700 - val_rmse: 3.5095\n",
"Epoch 788/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 1.3768 - mae: 1.7436 - mape: 1.1242 - r2: 0.8478 - rmse: 4.1231 - val_loss: 1.0945 - val_mae: 1.4376 - val_mape: 0.9365 - val_r2: 0.8744 - val_rmse: 3.4500\n",
"Epoch 789/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step - loss: 1.3934 - mae: 1.7614 - mape: 1.1363 - r2: 0.8386 - rmse: 4.2236 - val_loss: 1.0878 - val_mae: 1.4005 - val_mape: 0.9090 - val_r2: 0.8660 - val_rmse: 3.5625\n",
"Epoch 790/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 1.3929 - mae: 1.7613 - mape: 1.1335 - r2: 0.8393 - rmse: 4.2159 - val_loss: 1.1387 - val_mae: 1.4558 - val_mape: 0.9573 - val_r2: 0.8571 - val_rmse: 3.6790\n",
"Epoch 791/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 1.3421 - mae: 1.7083 - mape: 1.1023 - r2: 0.8537 - rmse: 4.0465 - val_loss: 1.1339 - val_mae: 1.4483 - val_mape: 0.9399 - val_r2: 0.8524 - val_rmse: 3.7387\n",
"Epoch 792/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 32ms/step - loss: 1.3671 - mae: 1.7339 - mape: 1.1185 - r2: 0.8399 - rmse: 4.1781 - val_loss: 1.1029 - val_mae: 1.4452 - val_mape: 0.9402 - val_r2: 0.8811 - val_rmse: 3.3559\n",
"Epoch 793/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 31ms/step - loss: 1.3887 - mae: 1.7554 - mape: 1.1342 - r2: 0.8401 - rmse: 4.2176 - val_loss: 1.1369 - val_mae: 1.4496 - val_mape: 0.9467 - val_r2: 0.8610 - val_rmse: 3.6292\n",
"Epoch 794/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 36ms/step - loss: 1.3678 - mae: 1.7344 - mape: 1.1168 - r2: 0.8457 - rmse: 4.1608 - val_loss: 1.0518 - val_mae: 1.3591 - val_mape: 0.8757 - val_r2: 0.8635 - val_rmse: 3.5955\n",
"Epoch 795/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 30ms/step - loss: 1.3617 - mae: 1.7262 - mape: 1.1145 - r2: 0.8419 - rmse: 4.2114 - val_loss: 1.1017 - val_mae: 1.4198 - val_mape: 0.9250 - val_r2: 0.8625 - val_rmse: 3.6094\n",
"Epoch 796/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 30ms/step - loss: 1.4172 - mae: 1.7804 - mape: 1.1515 - r2: 0.8332 - rmse: 4.3217 - val_loss: 1.0269 - val_mae: 1.3324 - val_mape: 0.8618 - val_r2: 0.8811 - val_rmse: 3.3560\n",
"Epoch 797/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 32ms/step - loss: 1.3808 - mae: 1.7440 - mape: 1.1271 - r2: 0.8381 - rmse: 4.2491 - val_loss: 1.1104 - val_mae: 1.4603 - val_mape: 0.9550 - val_r2: 0.8666 - val_rmse: 3.5546\n",
"Epoch 798/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step - loss: 1.3478 - mae: 1.7119 - mape: 1.1042 - r2: 0.8503 - rmse: 4.0935 - val_loss: 1.0999 - val_mae: 1.4338 - val_mape: 0.9316 - val_r2: 0.8672 - val_rmse: 3.5471\n",
"Epoch 799/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 35ms/step - loss: 1.3894 - mae: 1.7557 - mape: 1.1350 - r2: 0.8358 - rmse: 4.2467 - val_loss: 1.0999 - val_mae: 1.4377 - val_mape: 0.9425 - val_r2: 0.8710 - val_rmse: 3.4960\n",
"Epoch 800/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 39ms/step - loss: 1.3878 - mae: 1.7541 - mape: 1.1358 - r2: 0.8415 - rmse: 4.2112 - val_loss: 1.0578 - val_mae: 1.3606 - val_mape: 0.8903 - val_r2: 0.8783 - val_rmse: 3.3948\n",
"Epoch 801/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 38ms/step - loss: 1.3527 - mae: 1.7182 - mape: 1.1101 - r2: 0.8451 - rmse: 4.1627 - val_loss: 0.9941 - val_mae: 1.3055 - val_mape: 0.8447 - val_r2: 0.8825 - val_rmse: 3.3357\n",
"Epoch 802/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 1.3398 - mae: 1.7040 - mape: 1.0991 - r2: 0.8519 - rmse: 4.0604 - val_loss: 1.0479 - val_mae: 1.3705 - val_mape: 0.8955 - val_r2: 0.8803 - val_rmse: 3.3674\n",
"Epoch 803/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 37ms/step - loss: 1.3528 - mae: 1.7160 - mape: 1.1074 - r2: 0.8466 - rmse: 4.1375 - val_loss: 1.0814 - val_mae: 1.4006 - val_mape: 0.9056 - val_r2: 0.8730 - val_rmse: 3.4688\n",
"Epoch 804/2000\n",
"\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 34ms/step - loss: 1.3413 - mae: 1.7026 - mape: 1.0968 - r2: 0.8427 - rmse: 4.1403 - val_loss: 1.1301 - val_mae: 1.4466 - val_mape: 0.9463 - val_r2: 0.8640 - val_rmse: 3.5888\n",
"Epoch 804: early stopping\n",
"Restoring model weights from the end of the best epoch: 784.\n"
]
}
],
"source": [
"# ============================================================\n",
"# 9. TRAINING WITH COSINE LR SCHEDULE + EARLY STOPPING\n",
"# ============================================================\n",
"\n",
"\n",
"\n",
"# Cosine decay with linear warmup\n",
"class WarmupCosineDecay(tf.keras.optimizers.schedules.LearningRateSchedule):\n",
" def __init__(self, init_lr, warmup_steps, total_steps):\n",
" super().__init__()\n",
" self.init_lr = init_lr\n",
" self.warmup_steps = tf.cast(warmup_steps, tf.float32)\n",
" self.total_steps = tf.cast(total_steps, tf.float32)\n",
"\n",
" def __call__(self, step):\n",
" step = tf.cast(step, tf.float32)\n",
" warmup_lr = self.init_lr * (step / self.warmup_steps)\n",
" cosine_lr = self.init_lr * 0.5 * (\n",
" 1.0 + tf.math.cos(\n",
" np.pi * (step - self.warmup_steps) / (self.total_steps - self.warmup_steps)\n",
" )\n",
" )\n",
" return tf.where(step < self.warmup_steps, warmup_lr, cosine_lr)\n",
"\n",
" def get_config(self):\n",
" return {\"init_lr\": self.init_lr,\n",
" \"warmup_steps\": int(self.warmup_steps.numpy()),\n",
" \"total_steps\": int(self.total_steps.numpy())}\n",
"\n",
"class R2Score(tf.keras.metrics.Metric):\n",
" def __init__(self, name=\"r2\", **kwargs):\n",
" super().__init__(name=name, **kwargs)\n",
" self.sse = self.add_weight(name=\"sse\", initializer=\"zeros\")\n",
" self.sst = self.add_weight(name=\"sst\", initializer=\"zeros\")\n",
"\n",
" def update_state(self, y_true, y_pred, sample_weight=None):\n",
" y_true = tf.cast(y_true, tf.float32)\n",
" y_pred = tf.cast(y_pred, tf.float32)\n",
"\n",
" self.sse.assign_add(tf.reduce_sum(tf.square(y_true - y_pred)))\n",
" self.sst.assign_add(tf.reduce_sum(tf.square(y_true - tf.reduce_mean(y_true))))\n",
"\n",
" def result(self):\n",
" return 1 - (self.sse / (self.sst + 1e-8))\n",
"\n",
" def reset_state(self):\n",
" self.sse.assign(0.0)\n",
" self.sst.assign(0.0)\n",
"\n",
"EPOCHS = 2000\n",
"STEPS_TRAIN = len(hr_train) // BATCH_SIZE\n",
"\n",
"lr_schedule = WarmupCosineDecay(\n",
" init_lr=3e-4,\n",
" warmup_steps=5 * STEPS_TRAIN,\n",
" total_steps=EPOCHS * STEPS_TRAIN,\n",
")\n",
"\n",
"pretrain_model.compile(\n",
" optimizer=tf.keras.optimizers.Adam(lr_schedule),\n",
" loss=tf.keras.losses.Huber(),\n",
" metrics=[\n",
" tf.keras.metrics.MeanAbsoluteError(name=\"mae\"),\n",
" tf.keras.metrics.MeanAbsolutePercentageError(name=\"mape\"),\n",
" tf.keras.metrics.RootMeanSquaredError(name=\"rmse\"),\n",
" R2Score(),\n",
" ],\n",
")\n",
"\n",
"log_dir = os.path.join(\"logs\", \"pretrain\", datetime.datetime.now().strftime(\"%Y%m%d-%H%M%S\"), \"final\")\n",
"callbacks = [\n",
" tf.keras.callbacks.TensorBoard(log_dir=log_dir),\n",
" tf.keras.callbacks.EarlyStopping(\n",
" monitor=\"val_loss\", patience=20, restore_best_weights=True, verbose=1\n",
" ),\n",
"]\n",
"\n",
"history = pretrain_model.fit(\n",
" train_ds,\n",
" epochs=EPOCHS,\n",
" validation_data=val_ds,\n",
" callbacks=callbacks,\n",
")\n"
]
},
{
"cell_type": "code",
"execution_count": 11,
"id": "cell_08_save",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"WARNING:absl:You are saving your model as an HDF5 file via `model.save()` or `keras.saving.save_model(model)`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')` or `keras.saving.save_model(model, 'my_model.keras')`. \n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Encoder saved to encoder_pretrained.h5\n",
"Saved hr_only_data_windows.npz\n"
]
}
],
"source": [
"encoder.save(\"../../models/encoder_pretrained.h5\")\n",
"print(\"Encoder saved to encoder_pretrained.h5\")\n",
"\n",
"np.savez(\n",
" \"../../db/hr_only_data_windows.npz\",\n",
" hr_arr=hr_arr,\n",
")\n",
"print(\"Saved hr_only_data_windows.npz\")"
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "cell_09_evaluate",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"I0000 00:00:1775488491.308504 217005 dot_merger.cc:481] Merging Dots in computation: a_inference_one_step_on_data_403975__.21\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Validation MAE: 1.3940 BPM\n",
"(HR mean=150.7 std=10.5)\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"import matplotlib.pyplot as plt\n",
"import numpy as np\n",
"import tensorflow as tf\n",
"\n",
"all_mae = []\n",
"for inputs, y_true in val_ds:\n",
" y_pred = pretrain_model.predict(inputs, verbose=0)\n",
" y_pred = tf.convert_to_tensor(y_pred)\n",
" mae = tf.reduce_mean(tf.abs(y_pred - y_true))\n",
" all_mae.append(mae.numpy())\n",
"\n",
"print(f\"Validation MAE: {np.mean(all_mae):.4f} BPM\")\n",
"print(f\"(HR mean={hr_arr.mean():.1f} std={hr_arr.std():.1f})\")\n",
"\n",
"for inputs, y_true in val_ds.take(1):\n",
" y_pred = pretrain_model.predict(inputs, verbose=0)[0]\n",
" y_true_seq = y_true.numpy()[0]\n",
"\n",
"t = np.arange(SEQ_LEN)\n",
"plt.figure(figsize=(10,4))\n",
"plt.plot(t, y_true_seq, label=\"True HR\")\n",
"plt.plot(t, y_pred, \"--\", label=\"Pred HR\")\n",
"plt.legend()\n",
"plt.grid(True)\n",
"plt.tight_layout()\n",
"plt.savefig(\"pretrain_results.png\", dpi=120)\n",
"plt.show()"
]
},
{
"cell_type": "code",
"execution_count": 9,
"id": "e3b0fac2",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Masked MAE: 5.408 BPM\n",
"Masked Median: 3.521 BPM\n",
"Masked Std: 5.872 BPM\n",
"Masked 90th p: 11.704 BPM\n"
]
}
],
"source": [
"# Rebuild a val_ds_with_mask for eval\n",
"val_ds_with_mask = (\n",
" make_ds(hr_val, shuffle=False)\n",
" .map(add_mask, num_parallel_calls=tf.data.AUTOTUNE)\n",
")\n",
"\n",
"all_abs_errors = []\n",
"\n",
"for batch in val_ds_with_mask:\n",
" inputs = {\"hr\": batch[\"hr\"]}\n",
" y_true = batch[\"target\"] # (B, SEQ_LEN)\n",
" mask = batch[\"mask\"] # (B, SEQ_LEN)\n",
"\n",
" y_pred = pretrain_model.predict(inputs, verbose=0)\n",
" y_pred = tf.convert_to_tensor(y_pred)\n",
"\n",
" abs_errors = tf.abs(y_pred - y_true) * mask\n",
" all_abs_errors.append(abs_errors.numpy().flatten())\n",
"\n",
"all_abs_errors = np.concatenate(all_abs_errors)\n",
"masked_errors = all_abs_errors[all_abs_errors > 0]\n",
"\n",
"print(f\"Masked MAE: {masked_errors.mean():.3f} BPM\")\n",
"print(f\"Masked Median: {np.median(masked_errors):.3f} BPM\")\n",
"print(f\"Masked Std: {masked_errors.std():.3f} BPM\")\n",
"print(f\"Masked 90th p: {np.percentile(masked_errors, 90):.3f} BPM\")\n"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "3.12.10",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.12.10"
}
},
"nbformat": 4,
"nbformat_minor": 5
}