| { | |
| "best_metric": 0.8714285714285714, | |
| "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": 6.447523593902588, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 2.3807, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.13513513513513514, | |
| "grad_norm": 7.273040771484375, | |
| "learning_rate": 4.81203007518797e-05, | |
| "loss": 2.2314, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.20270270270270271, | |
| "grad_norm": 8.625953674316406, | |
| "learning_rate": 4.43609022556391e-05, | |
| "loss": 2.1403, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.25675675675675674, | |
| "eval_accuracy": 0.5142857142857142, | |
| "eval_loss": 1.731622338294983, | |
| "eval_runtime": 19.4919, | |
| "eval_samples_per_second": 3.591, | |
| "eval_steps_per_second": 0.462, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.0135135135135136, | |
| "grad_norm": 8.909628868103027, | |
| "learning_rate": 4.0601503759398494e-05, | |
| "loss": 1.968, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.0810810810810811, | |
| "grad_norm": 10.394527435302734, | |
| "learning_rate": 3.6842105263157895e-05, | |
| "loss": 1.5202, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.1486486486486487, | |
| "grad_norm": 11.732996940612793, | |
| "learning_rate": 3.3082706766917295e-05, | |
| "loss": 1.2272, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.2162162162162162, | |
| "grad_norm": 11.313809394836426, | |
| "learning_rate": 2.9323308270676693e-05, | |
| "loss": 0.8363, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.2567567567567568, | |
| "eval_accuracy": 0.7142857142857143, | |
| "eval_loss": 0.878302276134491, | |
| "eval_runtime": 19.5002, | |
| "eval_samples_per_second": 3.59, | |
| "eval_steps_per_second": 0.462, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.027027027027027, | |
| "grad_norm": 5.413862705230713, | |
| "learning_rate": 2.556390977443609e-05, | |
| "loss": 0.8147, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.0945945945945947, | |
| "grad_norm": 4.164318561553955, | |
| "learning_rate": 2.1804511278195487e-05, | |
| "loss": 0.578, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.1621621621621623, | |
| "grad_norm": 9.895386695861816, | |
| "learning_rate": 1.8045112781954888e-05, | |
| "loss": 0.4988, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.22972972972973, | |
| "grad_norm": 7.693885803222656, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 0.3773, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.2567567567567566, | |
| "eval_accuracy": 0.8142857142857143, | |
| "eval_loss": 0.4005982279777527, | |
| "eval_runtime": 19.7694, | |
| "eval_samples_per_second": 3.541, | |
| "eval_steps_per_second": 0.455, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 3.0405405405405403, | |
| "grad_norm": 2.8522019386291504, | |
| "learning_rate": 1.0526315789473684e-05, | |
| "loss": 0.3401, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.108108108108108, | |
| "grad_norm": 4.198346138000488, | |
| "learning_rate": 6.766917293233083e-06, | |
| "loss": 0.2326, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.175675675675676, | |
| "grad_norm": 7.3412394523620605, | |
| "learning_rate": 3.007518796992481e-06, | |
| "loss": 0.2545, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "eval_accuracy": 0.8714285714285714, | |
| "eval_loss": 0.3345661759376526, | |
| "eval_runtime": 21.3549, | |
| "eval_samples_per_second": 3.278, | |
| "eval_steps_per_second": 0.421, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "step": 148, | |
| "total_flos": 1.460491890402263e+18, | |
| "train_loss": 1.0534377798840806, | |
| "train_runtime": 801.9153, | |
| "train_samples_per_second": 1.476, | |
| "train_steps_per_second": 0.185 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "eval_accuracy": 0.8709677419354839, | |
| "eval_loss": 0.3756483495235443, | |
| "eval_runtime": 43.3016, | |
| "eval_samples_per_second": 3.58, | |
| "eval_steps_per_second": 0.462, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "eval_accuracy": 0.8709677419354839, | |
| "eval_loss": 0.37564828991889954, | |
| "eval_runtime": 43.6224, | |
| "eval_samples_per_second": 3.553, | |
| "eval_steps_per_second": 0.458, | |
| "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 | |
| } | |