| { | |
| "best_metric": 0.8627450980392157, | |
| "best_model_checkpoint": "vit-base-patch16-224-U8-10b\\checkpoint-160", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 5.5e-05, | |
| "loss": 1.3694, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 5.210526315789474e-05, | |
| "loss": 1.2473, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5882352941176471, | |
| "eval_loss": 1.1670756340026855, | |
| "eval_runtime": 0.9807, | |
| "eval_samples_per_second": 52.002, | |
| "eval_steps_per_second": 2.039, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 4.921052631578948e-05, | |
| "loss": 1.1179, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.631578947368421e-05, | |
| "loss": 0.955, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6470588235294118, | |
| "eval_loss": 0.9392386078834534, | |
| "eval_runtime": 0.9523, | |
| "eval_samples_per_second": 53.557, | |
| "eval_steps_per_second": 2.1, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 4.342105263157895e-05, | |
| "loss": 0.8458, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.0526315789473684e-05, | |
| "loss": 0.735, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6862745098039216, | |
| "eval_loss": 0.7247381210327148, | |
| "eval_runtime": 0.8402, | |
| "eval_samples_per_second": 60.7, | |
| "eval_steps_per_second": 2.38, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 3.7631578947368425e-05, | |
| "loss": 0.6208, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 3.473684210526316e-05, | |
| "loss": 0.5341, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8235294117647058, | |
| "eval_loss": 0.5976751446723938, | |
| "eval_runtime": 0.8153, | |
| "eval_samples_per_second": 62.551, | |
| "eval_steps_per_second": 2.453, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 3.1842105263157895e-05, | |
| "loss": 0.4512, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 2.894736842105263e-05, | |
| "loss": 0.3864, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7450980392156863, | |
| "eval_loss": 0.6555898785591125, | |
| "eval_runtime": 0.8565, | |
| "eval_samples_per_second": 59.545, | |
| "eval_steps_per_second": 2.335, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "learning_rate": 2.605263157894737e-05, | |
| "loss": 0.3234, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 2.3157894736842103e-05, | |
| "loss": 0.2837, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7254901960784313, | |
| "eval_loss": 0.6780798435211182, | |
| "eval_runtime": 0.9582, | |
| "eval_samples_per_second": 53.226, | |
| "eval_steps_per_second": 2.087, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "learning_rate": 2.0263157894736842e-05, | |
| "loss": 0.2643, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 1.736842105263158e-05, | |
| "loss": 0.2332, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8431372549019608, | |
| "eval_loss": 0.5419377088546753, | |
| "eval_runtime": 0.8658, | |
| "eval_samples_per_second": 58.905, | |
| "eval_steps_per_second": 2.31, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "learning_rate": 1.4473684210526315e-05, | |
| "loss": 0.2328, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 1.1578947368421052e-05, | |
| "loss": 0.1974, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8627450980392157, | |
| "eval_loss": 0.5349088907241821, | |
| "eval_runtime": 0.7975, | |
| "eval_samples_per_second": 63.951, | |
| "eval_steps_per_second": 2.508, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "learning_rate": 8.68421052631579e-06, | |
| "loss": 0.1871, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 5.789473684210526e-06, | |
| "loss": 0.1857, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8235294117647058, | |
| "eval_loss": 0.5605680346488953, | |
| "eval_runtime": 0.8589, | |
| "eval_samples_per_second": 59.376, | |
| "eval_steps_per_second": 2.328, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "learning_rate": 2.894736842105263e-06, | |
| "loss": 0.1915, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.1907, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8431372549019608, | |
| "eval_loss": 0.48754531145095825, | |
| "eval_runtime": 0.8369, | |
| "eval_samples_per_second": 60.941, | |
| "eval_steps_per_second": 2.39, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 200, | |
| "total_flos": 1.9683318186491904e+18, | |
| "train_loss": 0.527636388540268, | |
| "train_runtime": 410.4734, | |
| "train_samples_per_second": 61.88, | |
| "train_steps_per_second": 0.487 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 1.9683318186491904e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |