Upload app.py

app.py

@@ -10,7 +10,6 @@ import tensorflow as tf
 import numpy as np
 from PIL import Image
 from huggingface_hub import hf_hub_download
-from datasets import load_dataset
 
 # ── Configuration ────────────────────────────────────────────────────────────
 MODEL_REPO = "Rishi2455/Human-Activity-Recognition"
@@ -30,12 +29,6 @@ ACTIVITY_EMOJI = {
     "Sitting": "🪑", "Sleeping": "😴", "Texting": "📱", "Using Laptop": "💻",
 }
 
-CLASS_NAMES_LOWER = [
-    "calling", "clapping", "cycling", "dancing", "drinking",
-    "eating", "fighting", "hugging", "laughing", "listening_to_music",
-    "running", "sitting", "sleeping", "texting", "using_laptop",
-]
-
 # ── Download & load model ───────────────────────────────────────────────────
 print("⬇️  Downloading model...")
 model_path = hf_hub_download(repo_id=MODEL_REPO, filename=MODEL_FILE)
@@ -43,30 +36,20 @@ print("🔧  Loading model...")
 model = tf.keras.models.load_model(model_path, compile=False)
 print("✅  Model loaded!")
 
-# ── Prepare example images from dataset (ALL 15 classes) ────────────────────
+# ── Example images (baked into the repo under examples/) ────────────────────
 EXAMPLE_DIR = "examples"
-os.makedirs(EXAMPLE_DIR, exist_ok=True)
-
-example_paths = []
-try:
-    print("📦  Loading example images from dataset...")
-    ds = load_dataset("Bingsu/Human_Action_Recognition", split="test")
-    # Grab one example per class — all 15 activities
-    all_classes = set(range(15))
-    seen = set()
-    for row in ds:
-        label = row["labels"]
-        if label in all_classes and label not in seen:
-            img = row["image"]
-            fname = f"{EXAMPLE_DIR}/{CLASS_NAMES_LOWER[label]}.jpg"
-            img.save(fname)
-            example_paths.append(fname)
-            seen.add(label)
-        if len(seen) == 15:
-            break
-    print(f"✅  Saved {len(example_paths)} example images (all 15 classes).")
-except Exception as e:
-    print(f"⚠️  Could not load examples: {e}")
+EXAMPLE_FILES = [
+    "calling.jpg", "clapping.jpg", "cycling.jpg", "dancing.jpg",
+    "drinking.jpg", "eating.jpg", "fighting.jpg", "hugging.jpg",
+    "laughing.jpg", "listening_to_music.jpg", "running.jpg",
+    "sitting.jpg", "sleeping.jpg", "texting.jpg", "using_laptop.jpg",
+]
+example_paths = [
+    os.path.join(EXAMPLE_DIR, f)
+    for f in EXAMPLE_FILES
+    if os.path.exists(os.path.join(EXAMPLE_DIR, f))
+]
+print(f"📸  Found {len(example_paths)} example images.")
 
 # ── Inference ────────────────────────────────────────────────────────────────
 def predict(pil_img: Image.Image) -> dict:
@@ -149,10 +132,10 @@ with gr.Blocks(
                 label="📊 Prediction Results",
             )
 
-    # Examples — all 15 activity classes
+    # Examples — all 15 activity classes, baked into the repo
     if example_paths:
         gr.Examples(
-            examples=sorted(example_paths),
+            examples=example_paths,
             inputs=image_input,
             outputs=label_output,
             fn=predict,