| { | |
| "best_metric": 0.8205128205128205, | |
| "best_model_checkpoint": "videomae-base-finetuned-sample_kine/checkpoint-45", | |
| "epoch": 9.035714285714286, | |
| "eval_steps": 500, | |
| "global_step": 140, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07142857142857142, | |
| "grad_norm": 3.850412607192993, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.7564, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.10714285714285714, | |
| "eval_accuracy": 0.6923076923076923, | |
| "eval_loss": 0.666049599647522, | |
| "eval_runtime": 17.3838, | |
| "eval_samples_per_second": 2.243, | |
| "eval_steps_per_second": 0.288, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.0357142857142858, | |
| "grad_norm": 4.889276027679443, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 0.6276, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.1071428571428572, | |
| "grad_norm": 11.487154006958008, | |
| "learning_rate": 4.3650793650793655e-05, | |
| "loss": 0.6614, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.1071428571428572, | |
| "eval_accuracy": 0.6923076923076923, | |
| "eval_loss": 0.5676791071891785, | |
| "eval_runtime": 17.4778, | |
| "eval_samples_per_second": 2.231, | |
| "eval_steps_per_second": 0.286, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.0714285714285716, | |
| "grad_norm": 4.425611972808838, | |
| "learning_rate": 3.968253968253968e-05, | |
| "loss": 0.5941, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.107142857142857, | |
| "eval_accuracy": 0.8205128205128205, | |
| "eval_loss": 0.5078679323196411, | |
| "eval_runtime": 17.0179, | |
| "eval_samples_per_second": 2.292, | |
| "eval_steps_per_second": 0.294, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 3.0357142857142856, | |
| "grad_norm": 4.037534713745117, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.428, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.107142857142857, | |
| "grad_norm": 5.8046040534973145, | |
| "learning_rate": 3.1746031746031745e-05, | |
| "loss": 0.3661, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.107142857142857, | |
| "eval_accuracy": 0.6923076923076923, | |
| "eval_loss": 0.6174768209457397, | |
| "eval_runtime": 17.3537, | |
| "eval_samples_per_second": 2.247, | |
| "eval_steps_per_second": 0.288, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.071428571428571, | |
| "grad_norm": 1.8920663595199585, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.3258, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.107142857142857, | |
| "eval_accuracy": 0.7435897435897436, | |
| "eval_loss": 1.1648730039596558, | |
| "eval_runtime": 17.2838, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.289, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 5.035714285714286, | |
| "grad_norm": 23.643049240112305, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 0.4292, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 5.107142857142857, | |
| "grad_norm": 9.202912330627441, | |
| "learning_rate": 1.984126984126984e-05, | |
| "loss": 0.5887, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 5.107142857142857, | |
| "eval_accuracy": 0.717948717948718, | |
| "eval_loss": 0.4697076380252838, | |
| "eval_runtime": 17.1039, | |
| "eval_samples_per_second": 2.28, | |
| "eval_steps_per_second": 0.292, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 6.071428571428571, | |
| "grad_norm": 7.702420711517334, | |
| "learning_rate": 1.5873015873015872e-05, | |
| "loss": 0.3907, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.107142857142857, | |
| "eval_accuracy": 0.6153846153846154, | |
| "eval_loss": 0.9873863458633423, | |
| "eval_runtime": 17.325, | |
| "eval_samples_per_second": 2.251, | |
| "eval_steps_per_second": 0.289, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 7.035714285714286, | |
| "grad_norm": 8.652463912963867, | |
| "learning_rate": 1.1904761904761905e-05, | |
| "loss": 0.193, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 7.107142857142857, | |
| "grad_norm": 1.5207479000091553, | |
| "learning_rate": 7.936507936507936e-06, | |
| "loss": 0.1948, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 7.107142857142857, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_loss": 0.9959497451782227, | |
| "eval_runtime": 17.4802, | |
| "eval_samples_per_second": 2.231, | |
| "eval_steps_per_second": 0.286, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 8.071428571428571, | |
| "grad_norm": 11.438685417175293, | |
| "learning_rate": 3.968253968253968e-06, | |
| "loss": 0.1424, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 8.107142857142858, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_loss": 1.1356935501098633, | |
| "eval_runtime": 16.5116, | |
| "eval_samples_per_second": 2.362, | |
| "eval_steps_per_second": 0.303, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 9.035714285714286, | |
| "grad_norm": 1.0725109577178955, | |
| "learning_rate": 0.0, | |
| "loss": 0.2198, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 9.035714285714286, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_loss": 1.1466848850250244, | |
| "eval_runtime": 17.0652, | |
| "eval_samples_per_second": 2.285, | |
| "eval_steps_per_second": 0.293, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 9.035714285714286, | |
| "step": 140, | |
| "total_flos": 1.3619485962926162e+18, | |
| "train_loss": 0.4227044131074633, | |
| "train_runtime": 879.8726, | |
| "train_samples_per_second": 1.273, | |
| "train_steps_per_second": 0.159 | |
| }, | |
| { | |
| "epoch": 9.035714285714286, | |
| "eval_accuracy": 0.8205128205128205, | |
| "eval_loss": 0.5078679919242859, | |
| "eval_runtime": 16.7803, | |
| "eval_samples_per_second": 2.324, | |
| "eval_steps_per_second": 0.298, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 9.035714285714286, | |
| "eval_accuracy": 0.8205128205128205, | |
| "eval_loss": 0.5078681111335754, | |
| "eval_runtime": 17.1173, | |
| "eval_samples_per_second": 2.278, | |
| "eval_steps_per_second": 0.292, | |
| "step": 140 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 140, | |
| "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": 1.3619485962926162e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |