beit-base-patch16-224-pt22k-ft22k-finetuned-FER2013CKPlus-7e-05-finetuned-SFEW-7e-05
/
trainer_state.json
| { | |
| "best_metric": 0.49596309111880044, | |
| "best_model_checkpoint": "beit-base-patch16-224-pt22k-ft22k-finetuned-FER2013CKPlus-7e-05-finetuned-SFEW-7e-05/checkpoint-126", | |
| "epoch": 9.96551724137931, | |
| "global_step": 140, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 4.9999999999999996e-05, | |
| "loss": 2.1509, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_accuracy": 0.37254901960784315, | |
| "eval_loss": 1.6920275688171387, | |
| "eval_runtime": 10.5629, | |
| "eval_samples_per_second": 82.079, | |
| "eval_steps_per_second": 2.651, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 6.666666666666666e-05, | |
| "loss": 1.6764, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_accuracy": 0.46943483275663206, | |
| "eval_loss": 1.5035290718078613, | |
| "eval_runtime": 15.6046, | |
| "eval_samples_per_second": 55.561, | |
| "eval_steps_per_second": 1.794, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 6.111111111111111e-05, | |
| "loss": 1.4471, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 5.5555555555555545e-05, | |
| "loss": 1.2723, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "eval_accuracy": 0.46943483275663206, | |
| "eval_loss": 1.5060927867889404, | |
| "eval_runtime": 10.6696, | |
| "eval_samples_per_second": 81.259, | |
| "eval_steps_per_second": 2.624, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 4.9999999999999996e-05, | |
| "loss": 1.1746, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "eval_accuracy": 0.4728950403690888, | |
| "eval_loss": 1.5421292781829834, | |
| "eval_runtime": 11.4458, | |
| "eval_samples_per_second": 75.748, | |
| "eval_steps_per_second": 2.446, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 4.444444444444444e-05, | |
| "loss": 1.153, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "learning_rate": 3.8888888888888884e-05, | |
| "loss": 0.9954, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "eval_accuracy": 0.4786620530565167, | |
| "eval_loss": 1.565744400024414, | |
| "eval_runtime": 10.8831, | |
| "eval_samples_per_second": 79.665, | |
| "eval_steps_per_second": 2.573, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "learning_rate": 3.333333333333333e-05, | |
| "loss": 1.0029, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "eval_accuracy": 0.4844290657439446, | |
| "eval_loss": 1.586708664894104, | |
| "eval_runtime": 10.6127, | |
| "eval_samples_per_second": 81.695, | |
| "eval_steps_per_second": 2.638, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 6.41, | |
| "learning_rate": 2.7777777777777772e-05, | |
| "loss": 0.9139, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "eval_accuracy": 0.48788927335640137, | |
| "eval_loss": 1.594279408454895, | |
| "eval_runtime": 10.6847, | |
| "eval_samples_per_second": 81.144, | |
| "eval_steps_per_second": 2.621, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "learning_rate": 2.222222222222222e-05, | |
| "loss": 0.9519, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "learning_rate": 1.6666666666666664e-05, | |
| "loss": 0.8335, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 7.97, | |
| "eval_accuracy": 0.48904267589388695, | |
| "eval_loss": 1.6002804040908813, | |
| "eval_runtime": 11.2046, | |
| "eval_samples_per_second": 77.379, | |
| "eval_steps_per_second": 2.499, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "learning_rate": 1.111111111111111e-05, | |
| "loss": 0.8382, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 8.97, | |
| "eval_accuracy": 0.49596309111880044, | |
| "eval_loss": 1.5628858804702759, | |
| "eval_runtime": 10.522, | |
| "eval_samples_per_second": 82.399, | |
| "eval_steps_per_second": 2.661, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 9.28, | |
| "learning_rate": 5.555555555555555e-06, | |
| "loss": 0.7749, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 9.97, | |
| "learning_rate": 0.0, | |
| "loss": 0.7169, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 9.97, | |
| "eval_accuracy": 0.4855824682814302, | |
| "eval_loss": 1.5772030353546143, | |
| "eval_runtime": 10.4436, | |
| "eval_samples_per_second": 83.017, | |
| "eval_steps_per_second": 2.681, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 9.97, | |
| "step": 140, | |
| "total_flos": 1.427863337526786e+18, | |
| "train_loss": 1.1358447994504657, | |
| "train_runtime": 1053.1568, | |
| "train_samples_per_second": 17.557, | |
| "train_steps_per_second": 0.133 | |
| } | |
| ], | |
| "max_steps": 140, | |
| "num_train_epochs": 10, | |
| "total_flos": 1.427863337526786e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |