| { | |
| "best_metric": 0.5737453699111938, | |
| "best_model_checkpoint": "./vit-molecul/checkpoint-160", | |
| "epoch": 20.0, | |
| "global_step": 160, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 2.85e-06, | |
| "loss": 0.723, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.61, | |
| "eval_f1": 0.6096486838154338, | |
| "eval_loss": 0.6790218949317932, | |
| "eval_runtime": 19.8848, | |
| "eval_samples_per_second": 5.029, | |
| "eval_steps_per_second": 0.101, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 2.7e-06, | |
| "loss": 0.6915, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.62, | |
| "eval_f1": 0.5924495924495925, | |
| "eval_loss": 0.6660856604576111, | |
| "eval_runtime": 20.4688, | |
| "eval_samples_per_second": 4.885, | |
| "eval_steps_per_second": 0.098, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 2.55e-06, | |
| "loss": 0.6689, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.69, | |
| "eval_f1": 0.6892230576441103, | |
| "eval_loss": 0.647020697593689, | |
| "eval_runtime": 20.1108, | |
| "eval_samples_per_second": 4.972, | |
| "eval_steps_per_second": 0.099, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 0.6517, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.64, | |
| "eval_f1": 0.6376811594202898, | |
| "eval_loss": 0.6355536580085754, | |
| "eval_runtime": 17.3784, | |
| "eval_samples_per_second": 5.754, | |
| "eval_steps_per_second": 0.115, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 2.25e-06, | |
| "loss": 0.6368, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.72, | |
| "eval_f1": 0.7198879551820727, | |
| "eval_loss": 0.6288898587226868, | |
| "eval_runtime": 19.3335, | |
| "eval_samples_per_second": 5.172, | |
| "eval_steps_per_second": 0.103, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 2.1e-06, | |
| "loss": 0.621, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.73, | |
| "eval_f1": 0.7293233082706766, | |
| "eval_loss": 0.621720552444458, | |
| "eval_runtime": 21.1008, | |
| "eval_samples_per_second": 4.739, | |
| "eval_steps_per_second": 0.095, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 1.95e-06, | |
| "loss": 0.6061, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.69, | |
| "eval_f1": 0.6862030569895738, | |
| "eval_loss": 0.6197251081466675, | |
| "eval_runtime": 20.06, | |
| "eval_samples_per_second": 4.985, | |
| "eval_steps_per_second": 0.1, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 1.8e-06, | |
| "loss": 0.5924, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.73, | |
| "eval_f1": 0.7293233082706766, | |
| "eval_loss": 0.6086930632591248, | |
| "eval_runtime": 18.4413, | |
| "eval_samples_per_second": 5.423, | |
| "eval_steps_per_second": 0.108, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 1.65e-06, | |
| "loss": 0.5767, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.72, | |
| "eval_f1": 0.7198879551820727, | |
| "eval_loss": 0.6002515554428101, | |
| "eval_runtime": 19.1541, | |
| "eval_samples_per_second": 5.221, | |
| "eval_steps_per_second": 0.104, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 1.5e-06, | |
| "loss": 0.5633, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.72, | |
| "eval_f1": 0.7195512820512822, | |
| "eval_loss": 0.5953283905982971, | |
| "eval_runtime": 18.6111, | |
| "eval_samples_per_second": 5.373, | |
| "eval_steps_per_second": 0.107, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 1.35e-06, | |
| "loss": 0.5491, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.72, | |
| "eval_f1": 0.7198879551820727, | |
| "eval_loss": 0.5884998440742493, | |
| "eval_runtime": 19.1596, | |
| "eval_samples_per_second": 5.219, | |
| "eval_steps_per_second": 0.104, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 1.2000000000000002e-06, | |
| "loss": 0.5351, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.71, | |
| "eval_f1": 0.7099709970997101, | |
| "eval_loss": 0.5868653655052185, | |
| "eval_runtime": 19.9846, | |
| "eval_samples_per_second": 5.004, | |
| "eval_steps_per_second": 0.1, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 1.05e-06, | |
| "loss": 0.5239, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.7, | |
| "eval_f1": 0.6995192307692307, | |
| "eval_loss": 0.5866873264312744, | |
| "eval_runtime": 18.0538, | |
| "eval_samples_per_second": 5.539, | |
| "eval_steps_per_second": 0.111, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 9e-07, | |
| "loss": 0.5118, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.71, | |
| "eval_f1": 0.7099709970997101, | |
| "eval_loss": 0.5804287195205688, | |
| "eval_runtime": 13.1833, | |
| "eval_samples_per_second": 7.585, | |
| "eval_steps_per_second": 0.152, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 7.5e-07, | |
| "loss": 0.502, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.71, | |
| "eval_f1": 0.7099709970997101, | |
| "eval_loss": 0.5751881003379822, | |
| "eval_runtime": 14.0771, | |
| "eval_samples_per_second": 7.104, | |
| "eval_steps_per_second": 0.142, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 6.000000000000001e-07, | |
| "loss": 0.4942, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.72, | |
| "eval_f1": 0.7198879551820727, | |
| "eval_loss": 0.5737959742546082, | |
| "eval_runtime": 9.472, | |
| "eval_samples_per_second": 10.557, | |
| "eval_steps_per_second": 0.211, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 4.5e-07, | |
| "loss": 0.4885, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.71, | |
| "eval_f1": 0.7085720028137876, | |
| "eval_loss": 0.5770530104637146, | |
| "eval_runtime": 12.9019, | |
| "eval_samples_per_second": 7.751, | |
| "eval_steps_per_second": 0.155, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 3.0000000000000004e-07, | |
| "loss": 0.4831, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.71, | |
| "eval_f1": 0.7085720028137876, | |
| "eval_loss": 0.5750694870948792, | |
| "eval_runtime": 10.5808, | |
| "eval_samples_per_second": 9.451, | |
| "eval_steps_per_second": 0.189, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 1.5000000000000002e-07, | |
| "loss": 0.4793, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.71, | |
| "eval_f1": 0.7085720028137876, | |
| "eval_loss": 0.5742676258087158, | |
| "eval_runtime": 12.6314, | |
| "eval_samples_per_second": 7.917, | |
| "eval_steps_per_second": 0.158, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.4774, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.71, | |
| "eval_f1": 0.7085720028137876, | |
| "eval_loss": 0.5737453699111938, | |
| "eval_runtime": 16.6943, | |
| "eval_samples_per_second": 5.99, | |
| "eval_steps_per_second": 0.12, | |
| "step": 160 | |
| } | |
| ], | |
| "max_steps": 160, | |
| "num_train_epochs": 20, | |
| "total_flos": 6.19935916916736e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |