| { | |
| "best_metric": 0.9, | |
| "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-114", | |
| "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": 6.274120330810547, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 2.3166, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.13513513513513514, | |
| "grad_norm": 7.531378269195557, | |
| "learning_rate": 4.81203007518797e-05, | |
| "loss": 2.1931, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.20270270270270271, | |
| "grad_norm": 7.633078098297119, | |
| "learning_rate": 4.43609022556391e-05, | |
| "loss": 2.111, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.25675675675675674, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.8015570640563965, | |
| "eval_runtime": 20.5537, | |
| "eval_samples_per_second": 3.406, | |
| "eval_steps_per_second": 0.438, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.0135135135135136, | |
| "grad_norm": 11.260424613952637, | |
| "learning_rate": 4.0601503759398494e-05, | |
| "loss": 1.9378, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.0810810810810811, | |
| "grad_norm": 9.42284870147705, | |
| "learning_rate": 3.6842105263157895e-05, | |
| "loss": 1.5007, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.1486486486486487, | |
| "grad_norm": 10.917902946472168, | |
| "learning_rate": 3.3082706766917295e-05, | |
| "loss": 1.2527, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.2162162162162162, | |
| "grad_norm": 6.6887102127075195, | |
| "learning_rate": 2.9323308270676693e-05, | |
| "loss": 0.9127, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.2567567567567568, | |
| "eval_accuracy": 0.7571428571428571, | |
| "eval_loss": 0.8593908548355103, | |
| "eval_runtime": 19.406, | |
| "eval_samples_per_second": 3.607, | |
| "eval_steps_per_second": 0.464, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.027027027027027, | |
| "grad_norm": 7.911215782165527, | |
| "learning_rate": 2.556390977443609e-05, | |
| "loss": 0.8361, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.0945945945945947, | |
| "grad_norm": 17.41940689086914, | |
| "learning_rate": 2.1804511278195487e-05, | |
| "loss": 0.6269, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.1621621621621623, | |
| "grad_norm": 14.612859725952148, | |
| "learning_rate": 1.8045112781954888e-05, | |
| "loss": 0.6656, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.22972972972973, | |
| "grad_norm": 5.549419403076172, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 0.4873, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.2567567567567566, | |
| "eval_accuracy": 0.9, | |
| "eval_loss": 0.4942115247249603, | |
| "eval_runtime": 19.9286, | |
| "eval_samples_per_second": 3.513, | |
| "eval_steps_per_second": 0.452, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 3.0405405405405403, | |
| "grad_norm": 3.9581756591796875, | |
| "learning_rate": 1.0526315789473684e-05, | |
| "loss": 0.3739, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.108108108108108, | |
| "grad_norm": 14.909950256347656, | |
| "learning_rate": 6.766917293233083e-06, | |
| "loss": 0.3, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.175675675675676, | |
| "grad_norm": 7.838873386383057, | |
| "learning_rate": 3.007518796992481e-06, | |
| "loss": 0.3481, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "eval_accuracy": 0.8857142857142857, | |
| "eval_loss": 0.39098191261291504, | |
| "eval_runtime": 19.9901, | |
| "eval_samples_per_second": 3.502, | |
| "eval_steps_per_second": 0.45, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "step": 148, | |
| "total_flos": 1.460491890402263e+18, | |
| "train_loss": 1.0857644306646812, | |
| "train_runtime": 945.9168, | |
| "train_samples_per_second": 1.252, | |
| "train_steps_per_second": 0.156 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "eval_accuracy": 0.8258064516129032, | |
| "eval_loss": 0.5008033514022827, | |
| "eval_runtime": 41.8041, | |
| "eval_samples_per_second": 3.708, | |
| "eval_steps_per_second": 0.478, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "eval_accuracy": 0.8258064516129032, | |
| "eval_loss": 0.5008033514022827, | |
| "eval_runtime": 39.3717, | |
| "eval_samples_per_second": 3.937, | |
| "eval_steps_per_second": 0.508, | |
| "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 | |
| } | |