Hugging Face's logo

ethan1115
/
mint_temp

Model card Files
xet
 Community
ethan1115 commited on
Commit
9c3403f
·
verified ·
1 Parent(s): 040204e

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. outputs/pep-affibody-ablation-fe/libB_pseudo_bf16/best_model.pt +3 -0
  2. outputs/pep-affibody-ablation-fe/libB_pseudo_bf16/config.json +262 -0
  3. outputs/pep-affibody-ablation-fe/libB_pseudo_bf16/last_model.pt +3 -0
  4. outputs/pep-affibody-ablation-fe/libB_pseudo_bf16/metrics.csv +0 -0
  5. outputs/pep-affibody-ablation-fe/libB_pseudo_bf16/metrics.json +0 -0
  6. outputs/pep-affibody-ablation-fe/libB_pseudo_bf16/summary.txt +407 -0
  7. outputs/pep-affibody-ablation-fe/libB_pseudo_bf16/train.log +0 -0
outputs/pep-affibody-ablation-fe/libB_pseudo_bf16/best_model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0d72b5aa5573b04611dde1bdb3c2815337b4ca3938b3e9077ae65267e3b67f1e
3
+ size 3261776405
outputs/pep-affibody-ablation-fe/libB_pseudo_bf16/config.json ADDED
@@ -0,0 +1,262 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "args": {
3
+ "config_path": "data/esm2_t33_650M_UR50D.json",
4
+ "checkpoint_path": "mint.ckpt",
5
+ "train_csv": "data/libB/train_fold_enrichment_freqmean_pseudo_1to1.csv",
6
+ "val_csv": [
7
+ "data/libB/val_fold_enrichment_freqmean_pseudo_1to1.csv",
8
+ "data/libB/val_fold_enrichment_freqmean_pseudo_1to1_v2.csv"
9
+ ],
10
+ "best_val_index": 0,
11
+ "train_limit": null,
12
+ "log_transform": true,
13
+ "use_multimer": true,
14
+ "unfreeze_last_n": 5,
15
+ "sep_chains": true,
16
+ "hidden_dim": 512,
17
+ "dropout": 0.2,
18
+ "bs": 256,
19
+ "num_epochs": 10,
20
+ "lr": 0.001,
21
+ "backbone_lr": 0.0001,
22
+ "loss_fn": "mse",
23
+ "huber_delta": 1.0,
24
+ "truncation_rate": 0.0,
25
+ "truncation_warmup_steps": 3000,
26
+ "consistency_col": null,
27
+ "truncation_consistency_bonus": 0.5,
28
+ "grad_clip": 1.0,
29
+ "val_interval": 100,
30
+ "mixed_precision": "bf16",
31
+ "gradient_accumulation_steps": 2,
32
+ "activation_checkpointing": false,
33
+ "lr_scheduler_type": "constant",
34
+ "warmup_steps": 0,
35
+ "lr_min": 0.0,
36
+ "lr_scheduler_kwargs": "{}",
37
+ "sample_weight_col": null,
38
+ "negative_weight": 1.0,
39
+ "use_binding_quality": false,
40
+ "bq_w_count": 0.7,
41
+ "bq_w_fe": 0.3,
42
+ "no_wandb": false,
43
+ "wandb_project": "pep-affibody-ablation-fe",
44
+ "wandb_entity": "protein_llm",
45
+ "wandb_run_name": "libB_pseudo_bf16",
46
+ "output_dir": "/home/pj25000082/ku50001421/mint/outputs/pep-affibody-ablation-fe/libB_pseudo_bf16",
47
+ "seed": 42,
48
+ "num_workers": 4,
49
+ "prefetch_factor": 2,
50
+ "device": "cuda"
51
+ },
52
+ "model_cfg": {
53
+ "no_progress_bar": false,
54
+ "log_interval": 100,
55
+ "log_format": "json",
56
+ "azureml_logging": false,
57
+ "seed": 1,
58
+ "cpu": false,
59
+ "tpu": false,
60
+ "bf16": false,
61
+ "memory_efficient_bf16": false,
62
+ "fp16": true,
63
+ "memory_efficient_fp16": false,
64
+ "fp16_no_flatten_grads": false,
65
+ "fp16_init_scale": 4,
66
+ "fp16_scale_tolerance": 0.0,
67
+ "on_cpu_convert_precision": false,
68
+ "min_loss_scale": 0.0001,
69
+ "amp": false,
70
+ "amp_batch_retries": 2,
71
+ "amp_init_scale": 128,
72
+ "empty_cache_freq": 0,
73
+ "all_gather_list_size": 16384,
74
+ "model_parallel_size": 1,
75
+ "profile": false,
76
+ "reset_logging": false,
77
+ "suppress_crashes": false,
78
+ "use_plasma_view": false,
79
+ "plasma_path": "/tmp/plasma",
80
+ "criterion": "masked_lm",
81
+ "optimizer": "adam",
82
+ "lr_scheduler": "polynomial_decay",
83
+ "scoring": "bleu",
84
+ "task": "p_masked_lm_cluster_resample",
85
+ "num_workers": 0,
86
+ "skip_invalid_size_inputs_valid_test": true,
87
+ "max_tokens": 1024,
88
+ "required_batch_size_multiple": 8,
89
+ "required_seq_len_multiple": 1,
90
+ "dataset_impl": "fasta",
91
+ "data_buffer_size": 10,
92
+ "train_subset": "train50",
93
+ "valid_subset": "valid50",
94
+ "ignore_unused_valid_subsets": true,
95
+ "validate_interval": 99999,
96
+ "validate_interval_updates": 0,
97
+ "validate_after_updates": 0,
98
+ "disable_validation": false,
99
+ "max_tokens_valid": 1024,
100
+ "curriculum": 0,
101
+ "gen_subset": "test",
102
+ "num_shards": 1,
103
+ "shard_id": 0,
104
+ "distributed_world_size": 512,
105
+ "distributed_num_procs": 8,
106
+ "distributed_rank": 0,
107
+ "distributed_backend": "nccl",
108
+ "distributed_port": 14490,
109
+ "device_id": 0,
110
+ "distributed_no_spawn": false,
111
+ "ddp_backend": "c10d",
112
+ "ddp_comm_hook": "none",
113
+ "bucket_cap_mb": 25,
114
+ "fix_batches_to_gpus": false,
115
+ "find_unused_parameters": false,
116
+ "fast_stat_sync": false,
117
+ "heartbeat_timeout": 3600,
118
+ "broadcast_buffers": false,
119
+ "slowmo_algorithm": "LocalSGD",
120
+ "localsgd_frequency": 3,
121
+ "nprocs_per_node": 8,
122
+ "pipeline_model_parallel": false,
123
+ "pipeline_chunks": 0,
124
+ "pipeline_checkpoint": "never",
125
+ "zero_sharding": "none",
126
+ "no_reshard_after_forward": false,
127
+ "fp32_reduce_scatter": false,
128
+ "cpu_offload": false,
129
+ "use_sharded_state": false,
130
+ "arch": "p_roberta_large",
131
+ "max_epoch": 500,
132
+ "max_update": 0,
133
+ "stop_time_hours": 0,
134
+ "clip_norm": 0.0,
135
+ "use_inf_norm": false,
136
+ "sentence_avg": false,
137
+ "update_freq": [
138
+ 4
139
+ ],
140
+ "lr": [
141
+ 0.0004
142
+ ],
143
+ "stop_min_lr": -1.0,
144
+ "use_bmuf": false,
145
+ "save_dir": "/fsx-protein/halilakin/checkpoints/33layer_lr_poly.sample_ur50_to_90.ngpu512",
146
+ "restore_file": "checkpoint_last.pt",
147
+ "reset_dataloader": false,
148
+ "reset_lr_scheduler": false,
149
+ "reset_meters": false,
150
+ "reset_optimizer": false,
151
+ "optimizer_overrides": "{}",
152
+ "save_interval": 99999,
153
+ "save_interval_updates": 10000,
154
+ "keep_interval_updates": -1,
155
+ "keep_interval_updates_pattern": -1,
156
+ "keep_last_epochs": -1,
157
+ "keep_best_checkpoints": -1,
158
+ "no_save": false,
159
+ "no_epoch_checkpoints": true,
160
+ "no_last_checkpoints": false,
161
+ "no_save_optimizer_state": false,
162
+ "best_checkpoint_metric": "loss",
163
+ "maximize_best_checkpoint_metric": false,
164
+ "patience": -1,
165
+ "checkpoint_suffix": "",
166
+ "checkpoint_shard_count": 1,
167
+ "load_checkpoint_on_all_dp_ranks": false,
168
+ "write_checkpoints_asynchronously": false,
169
+ "no_mid_epoch_validate": false,
170
+ "encoder_layerdrop": 0,
171
+ "quant_noise_pq": 0,
172
+ "quant_noise_pq_block_size": 8,
173
+ "quant_noise_scalar": 0,
174
+ "min_params_to_wrap": 100000000,
175
+ "data": "/fsx-protein/zhongkai/datasets/202104esm2/03_output",
176
+ "sample_break_mode": "eos",
177
+ "tokens_per_sample": 1024,
178
+ "mask_prob": 0.15,
179
+ "leave_unmasked_prob": 0.1,
180
+ "random_token_prob": 0.1,
181
+ "freq_weighted_replacement": false,
182
+ "mask_whole_words": false,
183
+ "mask_multiple_length": 1,
184
+ "mask_stdev": 0.0,
185
+ "shorten_method": "random_crop",
186
+ "shorten_data_split_list": "train50",
187
+ "num_batch_buckets": 0,
188
+ "cluster_resample_fasta_path": "/fsx-protein/zhongkai/datasets/202104esm2/03_output/uniref90.filtered.fasta",
189
+ "cluster_resample_seq_id": 90,
190
+ "cluster_resample_ur50_ur90_ur100_path": "/fsx-protein/zhongkai/datasets/202104esm2/01_inputs/ur50_ur90_ur100.no_ur_id_prefix.csv",
191
+ "adam_betas": "[0.9,0.98]",
192
+ "adam_eps": 1e-08,
193
+ "weight_decay": 0.01,
194
+ "use_old_adam": false,
195
+ "warmup_updates": 2000,
196
+ "end_learning_rate": 2e-05,
197
+ "power": 1.0,
198
+ "total_num_update": "450000",
199
+ "pad": 1,
200
+ "eos": 2,
201
+ "unk": 3,
202
+ "max_positions": 1024,
203
+ "activation_fn": "gelu",
204
+ "use_rotary_embeddings": true,
205
+ "encoder_normalize_after": true,
206
+ "preact_normalize": true,
207
+ "token_dropout": true,
208
+ "layer_norm_fp32": true,
209
+ "attention_dropout": 0.0,
210
+ "dropout": 0.0,
211
+ "activation_dropout": 0.0,
212
+ "encoder_attention_heads": 20,
213
+ "encoder_embed_dim": 1280,
214
+ "encoder_ffn_embed_dim": 5120,
215
+ "encoder_layers": 33,
216
+ "no_seed_provided": false,
217
+ "pooler_activation_fn": "tanh",
218
+ "pooler_dropout": 0.0,
219
+ "encoder_normalize_before": false,
220
+ "encoder_learned_pos": false,
221
+ "use_bert_init": false,
222
+ "checkpoint_transformer_block": false,
223
+ "checkpoint_activations": false,
224
+ "effective_attention": false,
225
+ "_name": "p_roberta_large",
226
+ "untie_weights_roberta": false
227
+ },
228
+ "environment": {
229
+ "python_version": "3.10.19 (main, Oct 21 2025, 16:43:05) [GCC 11.2.0]",
230
+ "torch_version": "2.10.0+cu128",
231
+ "cuda_available": true,
232
+ "cuda_device": "NVIDIA H100 80GB HBM3",
233
+ "wandb_available": true
234
+ },
235
+ "git": {
236
+ "git_commit": "unknown",
237
+ "git_dirty": null
238
+ },
239
+ "timestamp": "2026-02-15 19:29:35",
240
+ "dataset": {
241
+ "train_csv": "data/libB/train_fold_enrichment_freqmean_pseudo_1to1.csv",
242
+ "val_csv": [
243
+ "data/libB/val_fold_enrichment_freqmean_pseudo_1to1.csv",
244
+ "data/libB/val_fold_enrichment_freqmean_pseudo_1to1_v2.csv"
245
+ ],
246
+ "train_size": 1481332,
247
+ "train_positive_ratio": 0.5,
248
+ "train_count_mean": 2.6410122781388643,
249
+ "train_count_std": 16.249330698762655,
250
+ "train_count_max": 17196.0,
251
+ "val_0_csv": "data/libB/val_fold_enrichment_freqmean_pseudo_1to1.csv",
252
+ "val_0_size": 176,
253
+ "val_0_positive_ratio": 0.7670454545454546,
254
+ "val_0_count_mean": 84.94886363636364,
255
+ "val_0_count_std": 426.4907605080106,
256
+ "val_1_csv": "data/libB/val_fold_enrichment_freqmean_pseudo_1to1_v2.csv",
257
+ "val_1_size": 165,
258
+ "val_1_positive_ratio": 0.806060606060606,
259
+ "val_1_count_mean": 77.43636363636364,
260
+ "val_1_count_std": 427.8391721268735
261
+ }
262
+ }
outputs/pep-affibody-ablation-fe/libB_pseudo_bf16/last_model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5efd660a2b0ecdf5baed0e1bcf2f8295ade391911f2e7db97d14df8c9facaf4e
3
+ size 7524953661
outputs/pep-affibody-ablation-fe/libB_pseudo_bf16/metrics.csv ADDED
The diff for this file is too large to render. See raw diff
 
outputs/pep-affibody-ablation-fe/libB_pseudo_bf16/metrics.json ADDED
The diff for this file is too large to render. See raw diff
 
outputs/pep-affibody-ablation-fe/libB_pseudo_bf16/summary.txt ADDED
@@ -0,0 +1,407 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ============================================================
2
+ Experiment Summary
3
+ ============================================================
4
+
5
+ Output dir: /home/pj25000082/ku50001421/mint/outputs/pep-affibody-ablation-fe/libB_pseudo_bf16
6
+ Timestamp: 2026-02-16 18:48:31
7
+ Duration: 83951.4 seconds (1399.2 minutes)
8
+
9
+ --- Key Hyperparameters ---
10
+ lr: 0.001
11
+ backbone_lr: 0.0001
12
+ bs: 256
13
+ num_epochs: 10
14
+ unfreeze_last_n: 5
15
+ hidden_dim: 512
16
+ dropout: 0.2
17
+ log_transform: True
18
+ sep_chains: True
19
+ loss_fn: mse
20
+ huber_delta: 1.0
21
+ truncation_rate: 0.0
22
+ truncation_warmup_steps: 3000
23
+ truncation_consistency_bonus: 0.5
24
+ val_interval: 100
25
+ grad_clip: 1.0
26
+ seed: 42
27
+ mixed_precision: bf16
28
+ gradient_accumulation_steps: 2
29
+ lr_scheduler_type: constant
30
+ warmup_steps: 0
31
+ lr_min: 0.0
32
+ lr_scheduler_kwargs: {}
33
+ activation_checkpointing: False
34
+ num_workers: 4
35
+ prefetch_factor: 2
36
+ negative_weight: 1.0
37
+ use_binding_quality: False
38
+ bq_w_count: 0.7
39
+ bq_w_fe: 0.3
40
+
41
+ --- Best Validation Metrics ---
42
+ val_0_cs0_kendall: 0.2156
43
+ val_0_cs0_n: 60
44
+ val_0_cs0_pearson: 0.21365348994731903
45
+ val_0_cs0_r2: -4.1103
46
+ val_0_cs0_rmse: 0.9724
47
+ val_0_cs0_spearman: 0.2781
48
+ val_0_cs1_kendall: 0.3810
49
+ val_0_cs1_n: 14
50
+ val_0_cs1_pearson: 0.6624651551246643
51
+ val_0_cs1_r2: -0.4791
52
+ val_0_cs1_rmse: 0.5618
53
+ val_0_cs1_spearman: 0.4532
54
+ val_0_cs2_kendall: 0.5490
55
+ val_0_cs2_n: 14
56
+ val_0_cs2_pearson: 0.8056180477142334
57
+ val_0_cs2_r2: 0.4388
58
+ val_0_cs2_rmse: 0.9954
59
+ val_0_cs2_spearman: 0.6675
60
+ val_0_cs3_kendall: 0.6030
61
+ val_0_cs3_n: 88
62
+ val_0_cs3_pearson: 0.6629645824432373
63
+ val_0_cs3_r2: 0.1671
64
+ val_0_cs3_rmse: 1.3475
65
+ val_0_cs3_spearman: 0.7335
66
+ val_0_kendall: 0.7003
67
+ val_0_pearson: 0.8241990208625793
68
+ val_0_pr_auc: 0.9677
69
+ val_0_r2: 0.6435
70
+ val_0_r2_orig: -0.0026
71
+ val_0_rmse: 1.1551
72
+ val_0_rmse_orig: 425.8286
73
+ val_0_roc_auc: 0.8944
74
+ val_0_spearman: 0.8574
75
+ val_0_top20_precision: 1.0000
76
+ val_0_top50_precision: 1.0000
77
+ val_1_cs0_kendall: 0.1705
78
+ val_1_cs0_n: 56
79
+ val_1_cs0_pearson: 0.1607104390859604
80
+ val_1_cs0_r2: -4.2089
81
+ val_1_cs0_rmse: 1.0479
82
+ val_1_cs0_spearman: 0.2263
83
+ val_1_cs1_kendall: 0.3552
84
+ val_1_cs1_n: 19
85
+ val_1_cs1_pearson: 0.7219028472900391
86
+ val_1_cs1_r2: -1.4061
87
+ val_1_cs1_rmse: 0.6211
88
+ val_1_cs1_spearman: 0.4364
89
+ val_1_cs2_kendall: 0.6250
90
+ val_1_cs2_n: 12
91
+ val_1_cs2_pearson: 0.8879897594451904
92
+ val_1_cs2_r2: 0.6522
93
+ val_1_cs2_rmse: 0.8304
94
+ val_1_cs2_spearman: 0.7666
95
+ val_1_cs3_kendall: 0.6641
96
+ val_1_cs3_n: 78
97
+ val_1_cs3_pearson: 0.7661534547805786
98
+ val_1_cs3_r2: 0.4029
99
+ val_1_cs3_rmse: 1.2134
100
+ val_1_cs3_spearman: 0.8095
101
+ val_1_kendall: 0.6468
102
+ val_1_pearson: 0.8435776829719543
103
+ val_1_pr_auc: 0.9530
104
+ val_1_r2: 0.6715
105
+ val_1_r2_orig: 0.0056
106
+ val_1_rmse: 1.0785
107
+ val_1_rmse_orig: 425.3369
108
+ val_1_roc_auc: 0.8253
109
+ val_1_spearman: 0.8048
110
+ val_1_top20_precision: 1.0000
111
+ val_1_top50_precision: 1.0000
112
+
113
+ --- Best Test Metrics ---
114
+
115
+ --- Training History (val_spearman) ---
116
+ step 100 (epoch 1): val_spearman=0.4002 train_loss=0.4755
117
+ step 200 (epoch 1): val_spearman=0.3998 train_loss=0.3835
118
+ step 300 (epoch 1): val_spearman=0.3618 train_loss=0.3674
119
+ step 400 (epoch 1): val_spearman=0.4301 train_loss=0.3556
120
+ step 500 (epoch 1): val_spearman=0.5208 train_loss=0.3525
121
+ step 600 (epoch 1): val_spearman=0.5011 train_loss=0.3524
122
+ step 700 (epoch 1): val_spearman=0.5102 train_loss=0.3553
123
+ step 800 (epoch 1): val_spearman=0.5923 train_loss=0.3336
124
+ step 900 (epoch 1): val_spearman=0.5592 train_loss=0.3499
125
+ step 1000 (epoch 1): val_spearman=0.6022 train_loss=0.3309
126
+ step 1100 (epoch 1): val_spearman=0.6729 train_loss=0.3378
127
+ step 1200 (epoch 1): val_spearman=0.6790 train_loss=0.3366
128
+ step 1300 (epoch 1): val_spearman=0.6750 train_loss=0.3252
129
+ step 1400 (epoch 1): val_spearman=0.7067 train_loss=0.3283
130
+ step 1500 (epoch 1): val_spearman=0.6072 train_loss=0.3320
131
+ step 1600 (epoch 1): val_spearman=0.6993 train_loss=0.3289
132
+ step 1700 (epoch 1): val_spearman=0.6476 train_loss=0.3312
133
+ step 1800 (epoch 1): val_spearman=0.6780 train_loss=0.3237
134
+ step 1900 (epoch 1): val_spearman=0.6616 train_loss=0.3295
135
+ step 2000 (epoch 1): val_spearman=0.6771 train_loss=0.3270
136
+ step 2100 (epoch 1): val_spearman=0.7104 train_loss=0.3328
137
+ step 2200 (epoch 1): val_spearman=0.6848 train_loss=0.3235
138
+ step 2300 (epoch 1): val_spearman=0.7076 train_loss=0.3243
139
+ step 2400 (epoch 1): val_spearman=0.6631 train_loss=0.3258
140
+ step 2500 (epoch 1): val_spearman=0.7335 train_loss=0.3313
141
+ step 2600 (epoch 1): val_spearman=0.7135 train_loss=0.3256
142
+ step 2700 (epoch 1): val_spearman=0.7599 train_loss=0.3250
143
+ step 2800 (epoch 1): val_spearman=0.7327 train_loss=0.3251
144
+ step 2900 (epoch 2): val_spearman=0.7286 train_loss=0.3280
145
+ step 3000 (epoch 2): val_spearman=0.7045 train_loss=0.3226
146
+ step 3100 (epoch 2): val_spearman=0.7314 train_loss=0.3182
147
+ step 3200 (epoch 2): val_spearman=0.7498 train_loss=0.3126
148
+ step 3300 (epoch 2): val_spearman=0.7618 train_loss=0.3178
149
+ step 3400 (epoch 2): val_spearman=0.7497 train_loss=0.3211
150
+ step 3500 (epoch 2): val_spearman=0.7116 train_loss=0.3228
151
+ step 3600 (epoch 2): val_spearman=0.7906 train_loss=0.3160
152
+ step 3700 (epoch 2): val_spearman=0.7276 train_loss=0.3146
153
+ step 3800 (epoch 2): val_spearman=0.7192 train_loss=0.3203
154
+ step 3900 (epoch 2): val_spearman=0.7611 train_loss=0.3152
155
+ step 4000 (epoch 2): val_spearman=0.7465 train_loss=0.3132
156
+ step 4100 (epoch 2): val_spearman=0.7588 train_loss=0.3167
157
+ step 4200 (epoch 2): val_spearman=0.7728 train_loss=0.3180
158
+ step 4300 (epoch 2): val_spearman=0.7827 train_loss=0.3137
159
+ step 4400 (epoch 2): val_spearman=0.7532 train_loss=0.3171
160
+ step 4500 (epoch 2): val_spearman=0.7677 train_loss=0.3173
161
+ step 4600 (epoch 2): val_spearman=0.7805 train_loss=0.3223
162
+ step 4700 (epoch 2): val_spearman=0.7669 train_loss=0.3137
163
+ step 4800 (epoch 2): val_spearman=0.7809 train_loss=0.3150
164
+ step 4900 (epoch 2): val_spearman=0.7627 train_loss=0.3196
165
+ step 5000 (epoch 2): val_spearman=0.7679 train_loss=0.3153
166
+ step 5100 (epoch 2): val_spearman=0.7821 train_loss=0.3182
167
+ step 5200 (epoch 2): val_spearman=0.7567 train_loss=0.3119
168
+ step 5300 (epoch 2): val_spearman=0.7915 train_loss=0.3161
169
+ step 5400 (epoch 2): val_spearman=0.7890 train_loss=0.3131
170
+ step 5500 (epoch 2): val_spearman=0.7941 train_loss=0.3148
171
+ step 5600 (epoch 2): val_spearman=0.7873 train_loss=0.3165
172
+ step 5700 (epoch 2): val_spearman=0.8025 train_loss=0.3134
173
+ step 5800 (epoch 3): val_spearman=0.7950 train_loss=0.3151
174
+ step 5900 (epoch 3): val_spearman=0.8108 train_loss=0.3074
175
+ step 6000 (epoch 3): val_spearman=0.8236 train_loss=0.3037
176
+ step 6100 (epoch 3): val_spearman=0.8253 train_loss=0.3115
177
+ step 6200 (epoch 3): val_spearman=0.8008 train_loss=0.3036
178
+ step 6300 (epoch 3): val_spearman=0.8184 train_loss=0.3031
179
+ step 6400 (epoch 3): val_spearman=0.8036 train_loss=0.3127
180
+ step 6500 (epoch 3): val_spearman=0.8262 train_loss=0.3057
181
+ step 6600 (epoch 3): val_spearman=0.8286 train_loss=0.3057
182
+ step 6700 (epoch 3): val_spearman=0.8057 train_loss=0.3049
183
+ step 6800 (epoch 3): val_spearman=0.8081 train_loss=0.3065
184
+ step 6900 (epoch 3): val_spearman=0.8071 train_loss=0.3097
185
+ step 7000 (epoch 3): val_spearman=0.7703 train_loss=0.3047
186
+ step 7100 (epoch 3): val_spearman=0.7687 train_loss=0.3128
187
+ step 7200 (epoch 3): val_spearman=0.8192 train_loss=0.3081
188
+ step 7300 (epoch 3): val_spearman=0.7907 train_loss=0.3060
189
+ step 7400 (epoch 3): val_spearman=0.8335 train_loss=0.3078
190
+ step 7500 (epoch 3): val_spearman=0.8353 train_loss=0.3058
191
+ step 7600 (epoch 3): val_spearman=0.8262 train_loss=0.3030
192
+ step 7700 (epoch 3): val_spearman=0.7816 train_loss=0.3041
193
+ step 7800 (epoch 3): val_spearman=0.8260 train_loss=0.3092
194
+ step 7900 (epoch 3): val_spearman=0.8271 train_loss=0.3121
195
+ step 8000 (epoch 3): val_spearman=0.8045 train_loss=0.3060
196
+ step 8100 (epoch 3): val_spearman=0.8065 train_loss=0.3092
197
+ step 8200 (epoch 3): val_spearman=0.8127 train_loss=0.3056
198
+ step 8300 (epoch 3): val_spearman=0.8522 train_loss=0.3035
199
+ step 8400 (epoch 3): val_spearman=0.8279 train_loss=0.3032
200
+ step 8500 (epoch 3): val_spearman=0.7696 train_loss=0.3043
201
+ step 8600 (epoch 3): val_spearman=0.8019 train_loss=0.3091
202
+ step 8700 (epoch 4): val_spearman=0.8305 train_loss=0.3000
203
+ step 8800 (epoch 4): val_spearman=0.8327 train_loss=0.2937
204
+ step 8900 (epoch 4): val_spearman=0.8328 train_loss=0.2958
205
+ step 9000 (epoch 4): val_spearman=0.8410 train_loss=0.2934
206
+ step 9100 (epoch 4): val_spearman=0.8147 train_loss=0.2920
207
+ step 9200 (epoch 4): val_spearman=0.8393 train_loss=0.2939
208
+ step 9300 (epoch 4): val_spearman=0.8374 train_loss=0.2980
209
+ step 9400 (epoch 4): val_spearman=0.8148 train_loss=0.2965
210
+ step 9500 (epoch 4): val_spearman=0.8293 train_loss=0.2966
211
+ step 9600 (epoch 4): val_spearman=0.8338 train_loss=0.2936
212
+ step 9700 (epoch 4): val_spearman=0.8351 train_loss=0.2930
213
+ step 9800 (epoch 4): val_spearman=0.8428 train_loss=0.2960
214
+ step 9900 (epoch 4): val_spearman=0.8366 train_loss=0.2918
215
+ step 10000 (epoch 4): val_spearman=0.8328 train_loss=0.2919
216
+ step 10100 (epoch 4): val_spearman=0.7875 train_loss=0.2996
217
+ step 10200 (epoch 4): val_spearman=0.8348 train_loss=0.2947
218
+ step 10300 (epoch 4): val_spearman=0.8331 train_loss=0.2984
219
+ step 10400 (epoch 4): val_spearman=0.8280 train_loss=0.2997
220
+ step 10500 (epoch 4): val_spearman=0.8456 train_loss=0.2971
221
+ step 10600 (epoch 4): val_spearman=0.8506 train_loss=0.2946
222
+ step 10700 (epoch 4): val_spearman=0.8430 train_loss=0.2898
223
+ step 10800 (epoch 4): val_spearman=0.8138 train_loss=0.3017
224
+ step 10900 (epoch 4): val_spearman=0.8340 train_loss=0.2893
225
+ step 11000 (epoch 4): val_spearman=0.8224 train_loss=0.2953
226
+ step 11100 (epoch 4): val_spearman=0.8470 train_loss=0.2918
227
+ step 11200 (epoch 4): val_spearman=0.8494 train_loss=0.2945
228
+ step 11300 (epoch 4): val_spearman=0.8489 train_loss=0.2956
229
+ step 11400 (epoch 4): val_spearman=0.8205 train_loss=0.2984
230
+ step 11500 (epoch 4): val_spearman=0.8382 train_loss=0.2965
231
+ step 11600 (epoch 5): val_spearman=0.8409 train_loss=0.2880
232
+ step 11700 (epoch 5): val_spearman=0.8279 train_loss=0.2761
233
+ step 11800 (epoch 5): val_spearman=0.8502 train_loss=0.2739
234
+ step 11900 (epoch 5): val_spearman=0.8417 train_loss=0.2820
235
+ step 12000 (epoch 5): val_spearman=0.8297 train_loss=0.2757
236
+ step 12100 (epoch 5): val_spearman=0.8439 train_loss=0.2741
237
+ step 12200 (epoch 5): val_spearman=0.8294 train_loss=0.2787
238
+ step 12300 (epoch 5): val_spearman=0.8114 train_loss=0.2780
239
+ step 12400 (epoch 5): val_spearman=0.8258 train_loss=0.2758
240
+ step 12500 (epoch 5): val_spearman=0.8336 train_loss=0.2740
241
+ step 12600 (epoch 5): val_spearman=0.8418 train_loss=0.2790
242
+ step 12700 (epoch 5): val_spearman=0.8234 train_loss=0.2784
243
+ step 12800 (epoch 5): val_spearman=0.8309 train_loss=0.2772
244
+ step 12900 (epoch 5): val_spearman=0.8363 train_loss=0.2829
245
+ step 13000 (epoch 5): val_spearman=0.8387 train_loss=0.2722
246
+ step 13100 (epoch 5): val_spearman=0.8339 train_loss=0.2735
247
+ step 13200 (epoch 5): val_spearman=0.8509 train_loss=0.2766
248
+ step 13300 (epoch 5): val_spearman=0.8283 train_loss=0.2801
249
+ step 13400 (epoch 5): val_spearman=0.8264 train_loss=0.2779
250
+ step 13500 (epoch 5): val_spearman=0.8189 train_loss=0.2802
251
+ step 13600 (epoch 5): val_spearman=0.8364 train_loss=0.2863
252
+ step 13700 (epoch 5): val_spearman=0.8282 train_loss=0.2816
253
+ step 13800 (epoch 5): val_spearman=0.8332 train_loss=0.2771
254
+ step 13900 (epoch 5): val_spearman=0.8305 train_loss=0.2797
255
+ step 14000 (epoch 5): val_spearman=0.8279 train_loss=0.2781
256
+ step 14100 (epoch 5): val_spearman=0.8404 train_loss=0.2826
257
+ step 14200 (epoch 5): val_spearman=0.8544 train_loss=0.2747
258
+ step 14300 (epoch 5): val_spearman=0.8383 train_loss=0.2831
259
+ step 14400 (epoch 5): val_spearman=0.8369 train_loss=0.2850
260
+ step 14500 (epoch 6): val_spearman=0.8354 train_loss=0.2710
261
+ step 14600 (epoch 6): val_spearman=0.8300 train_loss=0.2508
262
+ step 14700 (epoch 6): val_spearman=0.8354 train_loss=0.2515
263
+ step 14800 (epoch 6): val_spearman=0.8400 train_loss=0.2499
264
+ step 14900 (epoch 6): val_spearman=0.8379 train_loss=0.2522
265
+ step 15000 (epoch 6): val_spearman=0.8232 train_loss=0.2493
266
+ step 15100 (epoch 6): val_spearman=0.8278 train_loss=0.2523
267
+ step 15200 (epoch 6): val_spearman=0.8269 train_loss=0.2551
268
+ step 15300 (epoch 6): val_spearman=0.8331 train_loss=0.2528
269
+ step 15400 (epoch 6): val_spearman=0.8324 train_loss=0.2549
270
+ step 15500 (epoch 6): val_spearman=0.8192 train_loss=0.2558
271
+ step 15600 (epoch 6): val_spearman=0.8158 train_loss=0.2544
272
+ step 15700 (epoch 6): val_spearman=0.8011 train_loss=0.2510
273
+ step 15800 (epoch 6): val_spearman=0.8387 train_loss=0.2553
274
+ step 15900 (epoch 6): val_spearman=0.8131 train_loss=0.2565
275
+ step 16000 (epoch 6): val_spearman=0.7877 train_loss=0.2558
276
+ step 16100 (epoch 6): val_spearman=0.7925 train_loss=0.2523
277
+ step 16200 (epoch 6): val_spearman=0.8176 train_loss=0.2547
278
+ step 16300 (epoch 6): val_spearman=0.8234 train_loss=0.2538
279
+ step 16400 (epoch 6): val_spearman=0.8266 train_loss=0.2546
280
+ step 16500 (epoch 6): val_spearman=0.7923 train_loss=0.2566
281
+ step 16600 (epoch 6): val_spearman=0.7765 train_loss=0.2552
282
+ step 16700 (epoch 6): val_spearman=0.8091 train_loss=0.2558
283
+ step 16800 (epoch 6): val_spearman=0.8167 train_loss=0.2600
284
+ step 16900 (epoch 6): val_spearman=0.8427 train_loss=0.2571
285
+ step 17000 (epoch 6): val_spearman=0.8231 train_loss=0.2560
286
+ step 17100 (epoch 6): val_spearman=0.8282 train_loss=0.2524
287
+ step 17200 (epoch 6): val_spearman=0.8178 train_loss=0.2491
288
+ step 17300 (epoch 6): val_spearman=0.8430 train_loss=0.2518
289
+ step 17400 (epoch 7): val_spearman=0.8298 train_loss=0.2417
290
+ step 17500 (epoch 7): val_spearman=0.8379 train_loss=0.2157
291
+ step 17600 (epoch 7): val_spearman=0.8314 train_loss=0.2148
292
+ step 17700 (epoch 7): val_spearman=0.8321 train_loss=0.2184
293
+ step 17800 (epoch 7): val_spearman=0.8276 train_loss=0.2160
294
+ step 17900 (epoch 7): val_spearman=0.8434 train_loss=0.2202
295
+ step 18000 (epoch 7): val_spearman=0.8198 train_loss=0.2173
296
+ step 18100 (epoch 7): val_spearman=0.8182 train_loss=0.2249
297
+ step 18200 (epoch 7): val_spearman=0.8574 train_loss=0.2230
298
+ step 18300 (epoch 7): val_spearman=0.7976 train_loss=0.2188
299
+ step 18400 (epoch 7): val_spearman=0.8164 train_loss=0.2223
300
+ step 18500 (epoch 7): val_spearman=0.8203 train_loss=0.2225
301
+ step 18600 (epoch 7): val_spearman=0.8036 train_loss=0.2179
302
+ step 18700 (epoch 7): val_spearman=0.8321 train_loss=0.2166
303
+ step 18800 (epoch 7): val_spearman=0.8247 train_loss=0.2284
304
+ step 18900 (epoch 7): val_spearman=0.8291 train_loss=0.2275
305
+ step 19000 (epoch 7): val_spearman=0.8247 train_loss=0.2256
306
+ step 19100 (epoch 7): val_spearman=0.8103 train_loss=0.2252
307
+ step 19200 (epoch 7): val_spearman=0.8200 train_loss=0.2215
308
+ step 19300 (epoch 7): val_spearman=0.7826 train_loss=0.2254
309
+ step 19400 (epoch 7): val_spearman=0.8030 train_loss=0.2262
310
+ step 19500 (epoch 7): val_spearman=0.7987 train_loss=0.2239
311
+ step 19600 (epoch 7): val_spearman=0.8278 train_loss=0.2294
312
+ step 19700 (epoch 7): val_spearman=0.8073 train_loss=0.2216
313
+ step 19800 (epoch 7): val_spearman=0.8380 train_loss=0.2222
314
+ step 19900 (epoch 7): val_spearman=0.8328 train_loss=0.2188
315
+ step 20000 (epoch 7): val_spearman=0.8235 train_loss=0.2234
316
+ step 20100 (epoch 7): val_spearman=0.8227 train_loss=0.2302
317
+ step 20200 (epoch 7): val_spearman=0.8375 train_loss=0.2271
318
+ step 20300 (epoch 8): val_spearman=0.8444 train_loss=0.2103
319
+ step 20400 (epoch 8): val_spearman=0.8245 train_loss=0.1798
320
+ step 20500 (epoch 8): val_spearman=0.8064 train_loss=0.1790
321
+ step 20600 (epoch 8): val_spearman=0.8196 train_loss=0.1853
322
+ step 20700 (epoch 8): val_spearman=0.8127 train_loss=0.1830
323
+ step 20800 (epoch 8): val_spearman=0.8184 train_loss=0.1805
324
+ step 20900 (epoch 8): val_spearman=0.8061 train_loss=0.1858
325
+ step 21000 (epoch 8): val_spearman=0.8104 train_loss=0.1858
326
+ step 21100 (epoch 8): val_spearman=0.8109 train_loss=0.1849
327
+ step 21200 (epoch 8): val_spearman=0.8162 train_loss=0.1916
328
+ step 21300 (epoch 8): val_spearman=0.8056 train_loss=0.1876
329
+ step 21400 (epoch 8): val_spearman=0.8211 train_loss=0.1870
330
+ step 21500 (epoch 8): val_spearman=0.8292 train_loss=0.1864
331
+ step 21600 (epoch 8): val_spearman=0.8322 train_loss=0.1862
332
+ step 21700 (epoch 8): val_spearman=0.8245 train_loss=0.1868
333
+ step 21800 (epoch 8): val_spearman=0.8224 train_loss=0.1895
334
+ step 21900 (epoch 8): val_spearman=0.8371 train_loss=0.1901
335
+ step 22000 (epoch 8): val_spearman=0.8341 train_loss=0.1872
336
+ step 22100 (epoch 8): val_spearman=0.8309 train_loss=0.1879
337
+ step 22200 (epoch 8): val_spearman=0.8356 train_loss=0.1898
338
+ step 22300 (epoch 8): val_spearman=0.8307 train_loss=0.1876
339
+ step 22400 (epoch 8): val_spearman=0.8190 train_loss=0.1863
340
+ step 22500 (epoch 8): val_spearman=0.8232 train_loss=0.1885
341
+ step 22600 (epoch 8): val_spearman=0.8472 train_loss=0.1899
342
+ step 22700 (epoch 8): val_spearman=0.8271 train_loss=0.1910
343
+ step 22800 (epoch 8): val_spearman=0.8417 train_loss=0.1898
344
+ step 22900 (epoch 8): val_spearman=0.8468 train_loss=0.1891
345
+ step 23000 (epoch 8): val_spearman=0.8162 train_loss=0.1929
346
+ step 23100 (epoch 8): val_spearman=0.8312 train_loss=0.1861
347
+ step 23200 (epoch 9): val_spearman=0.8483 train_loss=0.1706
348
+ step 23300 (epoch 9): val_spearman=0.8381 train_loss=0.1465
349
+ step 23400 (epoch 9): val_spearman=0.8448 train_loss=0.1463
350
+ step 23500 (epoch 9): val_spearman=0.8297 train_loss=0.1476
351
+ step 23600 (epoch 9): val_spearman=0.8427 train_loss=0.1479
352
+ step 23700 (epoch 9): val_spearman=0.8299 train_loss=0.1492
353
+ step 23800 (epoch 9): val_spearman=0.8290 train_loss=0.1512
354
+ step 23900 (epoch 9): val_spearman=0.8368 train_loss=0.1505
355
+ step 24000 (epoch 9): val_spearman=0.8186 train_loss=0.1532
356
+ step 24100 (epoch 9): val_spearman=0.8286 train_loss=0.1496
357
+ step 24200 (epoch 9): val_spearman=0.8415 train_loss=0.1507
358
+ step 24300 (epoch 9): val_spearman=0.8488 train_loss=0.1513
359
+ step 24400 (epoch 9): val_spearman=0.8498 train_loss=0.1497
360
+ step 24500 (epoch 9): val_spearman=0.8522 train_loss=0.1560
361
+ step 24600 (epoch 9): val_spearman=0.8513 train_loss=0.1490
362
+ step 24700 (epoch 9): val_spearman=0.8284 train_loss=0.1543
363
+ step 24800 (epoch 9): val_spearman=0.8234 train_loss=0.1542
364
+ step 24900 (epoch 9): val_spearman=0.8147 train_loss=0.1525
365
+ step 25000 (epoch 9): val_spearman=0.8215 train_loss=0.1543
366
+ step 25100 (epoch 9): val_spearman=0.8334 train_loss=0.1550
367
+ step 25200 (epoch 9): val_spearman=0.8295 train_loss=0.1565
368
+ step 25300 (epoch 9): val_spearman=0.8220 train_loss=0.1602
369
+ step 25400 (epoch 9): val_spearman=0.8320 train_loss=0.1590
370
+ step 25500 (epoch 9): val_spearman=0.8422 train_loss=0.1617
371
+ step 25600 (epoch 9): val_spearman=0.8221 train_loss=0.1546
372
+ step 25700 (epoch 9): val_spearman=0.8255 train_loss=0.1555
373
+ step 25800 (epoch 9): val_spearman=0.8234 train_loss=0.1589
374
+ step 25900 (epoch 9): val_spearman=0.8169 train_loss=0.1547
375
+ step 26000 (epoch 9): val_spearman=0.8078 train_loss=0.1585
376
+ step 26100 (epoch 10): val_spearman=0.8251 train_loss=0.1381
377
+ step 26200 (epoch 10): val_spearman=0.8006 train_loss=0.1157
378
+ step 26300 (epoch 10): val_spearman=0.8446 train_loss=0.1167
379
+ step 26400 (epoch 10): val_spearman=0.8497 train_loss=0.1171
380
+ step 26500 (epoch 10): val_spearman=0.8121 train_loss=0.1210
381
+ step 26600 (epoch 10): val_spearman=0.8385 train_loss=0.1152
382
+ step 26700 (epoch 10): val_spearman=0.8209 train_loss=0.1181
383
+ step 26800 (epoch 10): val_spearman=0.8127 train_loss=0.1227
384
+ step 26900 (epoch 10): val_spearman=0.7972 train_loss=0.1243
385
+ step 27000 (epoch 10): val_spearman=0.8212 train_loss=0.1225
386
+ step 27100 (epoch 10): val_spearman=0.8100 train_loss=0.1228
387
+ step 27200 (epoch 10): val_spearman=0.8089 train_loss=0.1275
388
+ step 27300 (epoch 10): val_spearman=0.8190 train_loss=0.1242
389
+ step 27400 (epoch 10): val_spearman=0.8336 train_loss=0.1245
390
+ step 27500 (epoch 10): val_spearman=0.8118 train_loss=0.1229
391
+ step 27600 (epoch 10): val_spearman=0.8238 train_loss=0.1256
392
+ step 27700 (epoch 10): val_spearman=0.8090 train_loss=0.1257
393
+ step 27800 (epoch 10): val_spearman=0.7854 train_loss=0.1272
394
+ step 27900 (epoch 10): val_spearman=0.8196 train_loss=0.1283
395
+ step 28000 (epoch 10): val_spearman=0.8283 train_loss=0.1280
396
+ step 28100 (epoch 10): val_spearman=0.8144 train_loss=0.1276
397
+ step 28200 (epoch 10): val_spearman=0.8124 train_loss=0.1256
398
+ step 28300 (epoch 10): val_spearman=0.8188 train_loss=0.1299
399
+ step 28400 (epoch 10): val_spearman=0.7993 train_loss=0.1271
400
+ step 28500 (epoch 10): val_spearman=0.8174 train_loss=0.1266
401
+ step 28600 (epoch 10): val_spearman=0.7845 train_loss=0.1284
402
+ step 28700 (epoch 10): val_spearman=0.8076 train_loss=0.1269
403
+ step 28800 (epoch 10): val_spearman=0.8182 train_loss=0.1328
404
+ step 28900 (epoch 10): val_spearman=0.8231 train_loss=0.1283
405
+ step 28930 (epoch 11): val_spearman=0.8038 train_loss=0.1253
406
+
407
+ ============================================================
outputs/pep-affibody-ablation-fe/libB_pseudo_bf16/train.log ADDED
The diff for this file is too large to render. See raw diff