| { | |
| "best_metric": 0.8857142857142857, | |
| "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": 10.97286605834961, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 2.3649, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.13513513513513514, | |
| "grad_norm": 7.316098690032959, | |
| "learning_rate": 4.81203007518797e-05, | |
| "loss": 2.2922, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.20270270270270271, | |
| "grad_norm": 7.5828447341918945, | |
| "learning_rate": 4.43609022556391e-05, | |
| "loss": 2.1259, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.25675675675675674, | |
| "eval_accuracy": 0.2714285714285714, | |
| "eval_loss": 1.9268096685409546, | |
| "eval_runtime": 30.5426, | |
| "eval_samples_per_second": 2.292, | |
| "eval_steps_per_second": 0.295, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.0135135135135136, | |
| "grad_norm": 10.012065887451172, | |
| "learning_rate": 4.0601503759398494e-05, | |
| "loss": 1.948, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.0810810810810811, | |
| "grad_norm": 9.466978073120117, | |
| "learning_rate": 3.6842105263157895e-05, | |
| "loss": 1.5111, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.1486486486486487, | |
| "grad_norm": 9.027811050415039, | |
| "learning_rate": 3.3082706766917295e-05, | |
| "loss": 1.2952, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.2162162162162162, | |
| "grad_norm": 11.65381908416748, | |
| "learning_rate": 2.9323308270676693e-05, | |
| "loss": 1.0429, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.2567567567567568, | |
| "eval_accuracy": 0.6428571428571429, | |
| "eval_loss": 1.0446727275848389, | |
| "eval_runtime": 34.6872, | |
| "eval_samples_per_second": 2.018, | |
| "eval_steps_per_second": 0.259, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.027027027027027, | |
| "grad_norm": 10.774154663085938, | |
| "learning_rate": 2.556390977443609e-05, | |
| "loss": 0.9026, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.0945945945945947, | |
| "grad_norm": 8.876453399658203, | |
| "learning_rate": 2.1804511278195487e-05, | |
| "loss": 0.7068, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.1621621621621623, | |
| "grad_norm": 7.738159656524658, | |
| "learning_rate": 1.8045112781954888e-05, | |
| "loss": 0.6668, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.22972972972973, | |
| "grad_norm": 7.152702331542969, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 0.5065, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.2567567567567566, | |
| "eval_accuracy": 0.7714285714285715, | |
| "eval_loss": 0.6279496550559998, | |
| "eval_runtime": 29.5278, | |
| "eval_samples_per_second": 2.371, | |
| "eval_steps_per_second": 0.305, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 3.0405405405405403, | |
| "grad_norm": 16.146060943603516, | |
| "learning_rate": 1.0526315789473684e-05, | |
| "loss": 0.3893, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.108108108108108, | |
| "grad_norm": 3.8572874069213867, | |
| "learning_rate": 6.766917293233083e-06, | |
| "loss": 0.3515, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.175675675675676, | |
| "grad_norm": 10.520626068115234, | |
| "learning_rate": 3.007518796992481e-06, | |
| "loss": 0.3014, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "eval_accuracy": 0.8857142857142857, | |
| "eval_loss": 0.4684544801712036, | |
| "eval_runtime": 30.762, | |
| "eval_samples_per_second": 2.276, | |
| "eval_steps_per_second": 0.293, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "step": 148, | |
| "total_flos": 1.460491890402263e+18, | |
| "train_loss": 1.1273625766908801, | |
| "train_runtime": 1284.4849, | |
| "train_samples_per_second": 0.922, | |
| "train_steps_per_second": 0.115 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "eval_accuracy": 0.8258064516129032, | |
| "eval_loss": 0.5380022525787354, | |
| "eval_runtime": 70.433, | |
| "eval_samples_per_second": 2.201, | |
| "eval_steps_per_second": 0.284, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "eval_accuracy": 0.8258064516129032, | |
| "eval_loss": 0.5380021929740906, | |
| "eval_runtime": 71.5071, | |
| "eval_samples_per_second": 2.168, | |
| "eval_steps_per_second": 0.28, | |
| "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 | |
| } | |