| { | |
| "best_metric": 0.023517148569226265, | |
| "best_model_checkpoint": "./models/camera-type/checkpoint-200", | |
| "epoch": 5.0, | |
| "global_step": 2530, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 9.604743083003953e-05, | |
| "loss": 0.0703, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 9.213438735177866e-05, | |
| "loss": 0.0064, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_accuracy": 0.9915611814345991, | |
| "eval_loss": 0.023517148569226265, | |
| "eval_runtime": 13.6256, | |
| "eval_samples_per_second": 86.968, | |
| "eval_steps_per_second": 10.935, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 8.818181818181818e-05, | |
| "loss": 0.015, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 8.426877470355731e-05, | |
| "loss": 0.0034, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_accuracy": 0.9940928270042194, | |
| "eval_loss": 0.03918765112757683, | |
| "eval_runtime": 26.2367, | |
| "eval_samples_per_second": 45.166, | |
| "eval_steps_per_second": 5.679, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 8.031620553359685e-05, | |
| "loss": 0.0036, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 7.636363636363637e-05, | |
| "loss": 0.0066, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "eval_accuracy": 0.9839662447257383, | |
| "eval_loss": 0.10105644911527634, | |
| "eval_runtime": 13.5269, | |
| "eval_samples_per_second": 87.603, | |
| "eval_steps_per_second": 11.015, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 7.24110671936759e-05, | |
| "loss": 0.0001, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 6.845849802371542e-05, | |
| "loss": 0.0, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "eval_accuracy": 0.9839662447257383, | |
| "eval_loss": 0.12266893684864044, | |
| "eval_runtime": 14.2674, | |
| "eval_samples_per_second": 83.057, | |
| "eval_steps_per_second": 10.443, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 6.450592885375494e-05, | |
| "loss": 0.0, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 6.055335968379446e-05, | |
| "loss": 0.0, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "eval_accuracy": 0.9839662447257383, | |
| "eval_loss": 0.12320054322481155, | |
| "eval_runtime": 13.4323, | |
| "eval_samples_per_second": 88.22, | |
| "eval_steps_per_second": 11.093, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 5.660079051383399e-05, | |
| "loss": 0.0, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 5.2648221343873516e-05, | |
| "loss": 0.0, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "eval_accuracy": 0.9839662447257383, | |
| "eval_loss": 0.143290176987648, | |
| "eval_runtime": 14.5978, | |
| "eval_samples_per_second": 81.177, | |
| "eval_steps_per_second": 10.207, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 4.8695652173913046e-05, | |
| "loss": 0.0, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 4.4743083003952576e-05, | |
| "loss": 0.0, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_accuracy": 0.9839662447257383, | |
| "eval_loss": 0.14161579310894012, | |
| "eval_runtime": 13.9736, | |
| "eval_samples_per_second": 84.803, | |
| "eval_steps_per_second": 10.663, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 4.079051383399209e-05, | |
| "loss": 0.0, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 3.6837944664031623e-05, | |
| "loss": 0.0, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "eval_accuracy": 0.9839662447257383, | |
| "eval_loss": 0.14079466462135315, | |
| "eval_runtime": 13.3148, | |
| "eval_samples_per_second": 88.999, | |
| "eval_steps_per_second": 11.191, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 3.288537549407115e-05, | |
| "loss": 0.0, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 2.8932806324110674e-05, | |
| "loss": 0.0, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "eval_accuracy": 0.9839662447257383, | |
| "eval_loss": 0.140061616897583, | |
| "eval_runtime": 13.4092, | |
| "eval_samples_per_second": 88.372, | |
| "eval_steps_per_second": 11.112, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 2.4980237154150198e-05, | |
| "loss": 0.0, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 2.1027667984189725e-05, | |
| "loss": 0.0, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "eval_accuracy": 0.9839662447257383, | |
| "eval_loss": 0.13938628137111664, | |
| "eval_runtime": 13.5631, | |
| "eval_samples_per_second": 87.369, | |
| "eval_steps_per_second": 10.986, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "learning_rate": 1.707509881422925e-05, | |
| "loss": 0.0, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 1.3122529644268775e-05, | |
| "loss": 0.0, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "eval_accuracy": 0.9839662447257383, | |
| "eval_loss": 0.13903304934501648, | |
| "eval_runtime": 14.2016, | |
| "eval_samples_per_second": 83.441, | |
| "eval_steps_per_second": 10.492, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "learning_rate": 9.1699604743083e-06, | |
| "loss": 0.0, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 5.217391304347826e-06, | |
| "loss": 0.0, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "eval_accuracy": 0.9839662447257383, | |
| "eval_loss": 0.13887113332748413, | |
| "eval_runtime": 13.5995, | |
| "eval_samples_per_second": 87.136, | |
| "eval_steps_per_second": 10.956, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "learning_rate": 1.2648221343873517e-06, | |
| "loss": 0.0, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 2530, | |
| "total_flos": 1.9586275919110656e+18, | |
| "train_loss": 0.004172653943921321, | |
| "train_runtime": 720.5262, | |
| "train_samples_per_second": 35.079, | |
| "train_steps_per_second": 3.511 | |
| } | |
| ], | |
| "max_steps": 2530, | |
| "num_train_epochs": 5, | |
| "total_flos": 1.9586275919110656e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |