| { | |
| "best_metric": 0.8704663212435233, | |
| "best_model_checkpoint": "./vit-focal-skin/checkpoint-1878", | |
| "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": 0.1663, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00018402555910543132, | |
| "loss": 0.1962, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.000176038338658147, | |
| "loss": 0.1853, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.00016805111821086263, | |
| "loss": 0.1706, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0001600638977635783, | |
| "loss": 0.1593, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.00015207667731629394, | |
| "loss": 0.1702, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8393782383419689, | |
| "eval_f1": 0.8380951982999452, | |
| "eval_loss": 0.39222732186317444, | |
| "eval_precision": 0.857762154943502, | |
| "eval_recall": 0.8393782383419689, | |
| "eval_runtime": 1.5181, | |
| "eval_samples_per_second": 127.129, | |
| "eval_steps_per_second": 16.467, | |
| "step": 626 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.00014408945686900958, | |
| "loss": 0.1036, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 0.00013610223642172525, | |
| "loss": 0.0941, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.00012819488817891373, | |
| "loss": 0.1029, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0001202076677316294, | |
| "loss": 0.095, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 0.00011222044728434504, | |
| "loss": 0.0911, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 0.00010423322683706072, | |
| "loss": 0.0647, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8238341968911918, | |
| "eval_f1": 0.8247834676883651, | |
| "eval_loss": 0.561523973941803, | |
| "eval_precision": 0.8404391615022521, | |
| "eval_recall": 0.8238341968911918, | |
| "eval_runtime": 1.446, | |
| "eval_samples_per_second": 133.47, | |
| "eval_steps_per_second": 17.289, | |
| "step": 1252 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 9.624600638977636e-05, | |
| "loss": 0.052, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 8.825878594249202e-05, | |
| "loss": 0.0298, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 8.027156549520767e-05, | |
| "loss": 0.0243, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 7.228434504792333e-05, | |
| "loss": 0.0332, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 6.429712460063898e-05, | |
| "loss": 0.0267, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 5.630990415335463e-05, | |
| "loss": 0.0111, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8704663212435233, | |
| "eval_f1": 0.868408303886561, | |
| "eval_loss": 0.4315575659275055, | |
| "eval_precision": 0.8670271155479244, | |
| "eval_recall": 0.8704663212435233, | |
| "eval_runtime": 1.4834, | |
| "eval_samples_per_second": 130.104, | |
| "eval_steps_per_second": 16.853, | |
| "step": 1878 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 4.832268370607029e-05, | |
| "loss": 0.0175, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 4.0335463258785946e-05, | |
| "loss": 0.0083, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 3.23482428115016e-05, | |
| "loss": 0.0038, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 2.4361022364217255e-05, | |
| "loss": 0.0009, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 1.6373801916932906e-05, | |
| "loss": 0.0034, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 8.386581469648563e-06, | |
| "loss": 0.0025, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 3.9936102236421723e-07, | |
| "loss": 0.0034, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8601036269430051, | |
| "eval_f1": 0.8616514102008719, | |
| "eval_loss": 0.4513249099254608, | |
| "eval_precision": 0.8650015326151586, | |
| "eval_recall": 0.8601036269430051, | |
| "eval_runtime": 1.498, | |
| "eval_samples_per_second": 128.838, | |
| "eval_steps_per_second": 16.689, | |
| "step": 2504 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 2504, | |
| "total_flos": 3.104468219559813e+18, | |
| "train_loss": 0.0725347773061683, | |
| "train_runtime": 849.1916, | |
| "train_samples_per_second": 47.174, | |
| "train_steps_per_second": 2.949 | |
| } | |
| ], | |
| "max_steps": 2504, | |
| "num_train_epochs": 4, | |
| "total_flos": 3.104468219559813e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |