| { | |
| "best_metric": 0.8428571428571429, | |
| "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-148", | |
| "epoch": 3.22972972972973, | |
| "eval_steps": 500, | |
| "global_step": 148, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06756756756756757, | |
| "grad_norm": 8.218301773071289, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 2.355, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.13513513513513514, | |
| "grad_norm": 7.125568389892578, | |
| "learning_rate": 4.81203007518797e-05, | |
| "loss": 2.2296, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.20270270270270271, | |
| "grad_norm": 7.601614475250244, | |
| "learning_rate": 4.43609022556391e-05, | |
| "loss": 2.1381, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.25675675675675674, | |
| "eval_accuracy": 0.5428571428571428, | |
| "eval_loss": 1.8478083610534668, | |
| "eval_runtime": 29.2478, | |
| "eval_samples_per_second": 2.393, | |
| "eval_steps_per_second": 0.308, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.0135135135135136, | |
| "grad_norm": 9.502254486083984, | |
| "learning_rate": 4.0601503759398494e-05, | |
| "loss": 1.9897, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.0810810810810811, | |
| "grad_norm": 9.743173599243164, | |
| "learning_rate": 3.6842105263157895e-05, | |
| "loss": 1.6464, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.1486486486486487, | |
| "grad_norm": 9.88233757019043, | |
| "learning_rate": 3.3082706766917295e-05, | |
| "loss": 1.3069, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.2162162162162162, | |
| "grad_norm": 7.763911247253418, | |
| "learning_rate": 2.9323308270676693e-05, | |
| "loss": 0.9543, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.2567567567567568, | |
| "eval_accuracy": 0.6571428571428571, | |
| "eval_loss": 0.9674364328384399, | |
| "eval_runtime": 29.3573, | |
| "eval_samples_per_second": 2.384, | |
| "eval_steps_per_second": 0.307, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.027027027027027, | |
| "grad_norm": 8.281584739685059, | |
| "learning_rate": 2.556390977443609e-05, | |
| "loss": 0.8358, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.0945945945945947, | |
| "grad_norm": 18.204240798950195, | |
| "learning_rate": 2.1804511278195487e-05, | |
| "loss": 0.6852, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.1621621621621623, | |
| "grad_norm": 10.050329208374023, | |
| "learning_rate": 1.8045112781954888e-05, | |
| "loss": 0.6309, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.22972972972973, | |
| "grad_norm": 4.531847953796387, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 0.4914, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.2567567567567566, | |
| "eval_accuracy": 0.7428571428571429, | |
| "eval_loss": 0.6266418695449829, | |
| "eval_runtime": 30.3275, | |
| "eval_samples_per_second": 2.308, | |
| "eval_steps_per_second": 0.297, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 3.0405405405405403, | |
| "grad_norm": 3.251669406890869, | |
| "learning_rate": 1.0526315789473684e-05, | |
| "loss": 0.42, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.108108108108108, | |
| "grad_norm": 3.1868138313293457, | |
| "learning_rate": 6.766917293233083e-06, | |
| "loss": 0.3493, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.175675675675676, | |
| "grad_norm": 9.545303344726562, | |
| "learning_rate": 3.007518796992481e-06, | |
| "loss": 0.3415, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "eval_accuracy": 0.8428571428571429, | |
| "eval_loss": 0.4506227374076843, | |
| "eval_runtime": 30.195, | |
| "eval_samples_per_second": 2.318, | |
| "eval_steps_per_second": 0.298, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "step": 148, | |
| "total_flos": 1.460491890402263e+18, | |
| "train_loss": 1.1224354151132945, | |
| "train_runtime": 3980.8054, | |
| "train_samples_per_second": 0.297, | |
| "train_steps_per_second": 0.037 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "eval_accuracy": 0.9354838709677419, | |
| "eval_loss": 0.39373940229415894, | |
| "eval_runtime": 67.6876, | |
| "eval_samples_per_second": 2.29, | |
| "eval_steps_per_second": 0.295, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "eval_accuracy": 0.9354838709677419, | |
| "eval_loss": 0.3937394320964813, | |
| "eval_runtime": 67.6788, | |
| "eval_samples_per_second": 2.29, | |
| "eval_steps_per_second": 0.296, | |
| "step": 148 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 148, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.460491890402263e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |