| { | |
| "best_metric": 0.8860103626943006, | |
| "best_model_checkpoint": "./vit-focal-skin/checkpoint-1252", | |
| "epoch": 4.0, | |
| "global_step": 2504, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00019201277955271565, | |
| "loss": 0.7905, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00018402555910543132, | |
| "loss": 0.6162, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.000176038338658147, | |
| "loss": 0.5709, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.00016805111821086263, | |
| "loss": 0.5101, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0001600638977635783, | |
| "loss": 0.4966, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.00015207667731629394, | |
| "loss": 0.4563, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8341968911917098, | |
| "eval_f1": 0.8331282144797781, | |
| "eval_loss": 0.40268635749816895, | |
| "eval_precision": 0.8791963514680484, | |
| "eval_recall": 0.8341968911917098, | |
| "eval_runtime": 1.5386, | |
| "eval_samples_per_second": 125.44, | |
| "eval_steps_per_second": 16.249, | |
| "step": 626 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.00014408945686900958, | |
| "loss": 0.385, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 0.00013610223642172525, | |
| "loss": 0.3621, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.00012811501597444092, | |
| "loss": 0.3624, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.00012012779552715656, | |
| "loss": 0.3403, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 0.00011214057507987221, | |
| "loss": 0.3293, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 0.00010415335463258787, | |
| "loss": 0.2954, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8860103626943006, | |
| "eval_f1": 0.8891954516071332, | |
| "eval_loss": 0.3063604235649109, | |
| "eval_precision": 0.8988400162775354, | |
| "eval_recall": 0.8860103626943006, | |
| "eval_runtime": 1.5206, | |
| "eval_samples_per_second": 126.921, | |
| "eval_steps_per_second": 16.441, | |
| "step": 1252 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 9.616613418530351e-05, | |
| "loss": 0.2258, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 8.817891373801918e-05, | |
| "loss": 0.1583, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 8.019169329073483e-05, | |
| "loss": 0.1589, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 7.220447284345049e-05, | |
| "loss": 0.1271, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 6.421725239616614e-05, | |
| "loss": 0.1746, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 5.623003194888179e-05, | |
| "loss": 0.1118, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8704663212435233, | |
| "eval_f1": 0.8663258953141536, | |
| "eval_loss": 0.30049628019332886, | |
| "eval_precision": 0.8698602729520133, | |
| "eval_recall": 0.8704663212435233, | |
| "eval_runtime": 1.4368, | |
| "eval_samples_per_second": 134.33, | |
| "eval_steps_per_second": 17.4, | |
| "step": 1878 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 4.824281150159744e-05, | |
| "loss": 0.1193, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 4.0255591054313104e-05, | |
| "loss": 0.0409, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 3.226837060702875e-05, | |
| "loss": 0.0327, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 2.428115015974441e-05, | |
| "loss": 0.0357, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 1.6293929712460065e-05, | |
| "loss": 0.0386, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 8.306709265175718e-06, | |
| "loss": 0.0303, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 3.194888178913738e-07, | |
| "loss": 0.0317, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8549222797927462, | |
| "eval_f1": 0.8560478324319817, | |
| "eval_loss": 0.3552953004837036, | |
| "eval_precision": 0.8595484172497833, | |
| "eval_recall": 0.8549222797927462, | |
| "eval_runtime": 1.5113, | |
| "eval_samples_per_second": 127.703, | |
| "eval_steps_per_second": 16.542, | |
| "step": 2504 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 2504, | |
| "total_flos": 3.104468219559813e+18, | |
| "train_loss": 0.2716429328141264, | |
| "train_runtime": 832.7663, | |
| "train_samples_per_second": 48.105, | |
| "train_steps_per_second": 3.007 | |
| } | |
| ], | |
| "max_steps": 2504, | |
| "num_train_epochs": 4, | |
| "total_flos": 3.104468219559813e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |