| { | |
| "best_metric": 0.8571428571428571, | |
| "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": 2.804513454437256, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 2.3196, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.13513513513513514, | |
| "grad_norm": 3.077538251876831, | |
| "learning_rate": 4.81203007518797e-05, | |
| "loss": 2.2828, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.20270270270270271, | |
| "grad_norm": 3.1245503425598145, | |
| "learning_rate": 4.43609022556391e-05, | |
| "loss": 2.2293, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.25675675675675674, | |
| "eval_accuracy": 0.4714285714285714, | |
| "eval_loss": 2.0989785194396973, | |
| "eval_runtime": 17.4922, | |
| "eval_samples_per_second": 4.002, | |
| "eval_steps_per_second": 0.515, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.0135135135135136, | |
| "grad_norm": 4.262564659118652, | |
| "learning_rate": 4.0601503759398494e-05, | |
| "loss": 2.1885, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.0810810810810811, | |
| "grad_norm": 4.817582130432129, | |
| "learning_rate": 3.6842105263157895e-05, | |
| "loss": 2.0441, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.1486486486486487, | |
| "grad_norm": 6.194489002227783, | |
| "learning_rate": 3.3082706766917295e-05, | |
| "loss": 1.8597, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.2162162162162162, | |
| "grad_norm": 4.391678333282471, | |
| "learning_rate": 2.9323308270676693e-05, | |
| "loss": 1.6003, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.2567567567567568, | |
| "eval_accuracy": 0.6714285714285714, | |
| "eval_loss": 1.554324746131897, | |
| "eval_runtime": 22.5133, | |
| "eval_samples_per_second": 3.109, | |
| "eval_steps_per_second": 0.4, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.027027027027027, | |
| "grad_norm": 5.160553932189941, | |
| "learning_rate": 2.556390977443609e-05, | |
| "loss": 1.5679, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.0945945945945947, | |
| "grad_norm": 6.264176368713379, | |
| "learning_rate": 2.1804511278195487e-05, | |
| "loss": 1.3448, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.1621621621621623, | |
| "grad_norm": 8.144826889038086, | |
| "learning_rate": 1.8045112781954888e-05, | |
| "loss": 1.2506, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.22972972972973, | |
| "grad_norm": 3.8042192459106445, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 1.1744, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.2567567567567566, | |
| "eval_accuracy": 0.8285714285714286, | |
| "eval_loss": 1.157400131225586, | |
| "eval_runtime": 24.0665, | |
| "eval_samples_per_second": 2.909, | |
| "eval_steps_per_second": 0.374, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 3.0405405405405403, | |
| "grad_norm": 4.986996173858643, | |
| "learning_rate": 1.0526315789473684e-05, | |
| "loss": 1.0651, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.108108108108108, | |
| "grad_norm": 4.7840962409973145, | |
| "learning_rate": 6.766917293233083e-06, | |
| "loss": 0.9311, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.175675675675676, | |
| "grad_norm": 7.304906845092773, | |
| "learning_rate": 3.007518796992481e-06, | |
| "loss": 0.9129, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_loss": 1.009927749633789, | |
| "eval_runtime": 19.312, | |
| "eval_samples_per_second": 3.625, | |
| "eval_steps_per_second": 0.466, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "step": 148, | |
| "total_flos": 1.4620505633437778e+18, | |
| "train_loss": 1.5866705855807743, | |
| "train_runtime": 1158.2809, | |
| "train_samples_per_second": 1.022, | |
| "train_steps_per_second": 0.128 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 1.0212879180908203, | |
| "eval_runtime": 43.7965, | |
| "eval_samples_per_second": 3.539, | |
| "eval_steps_per_second": 0.457, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 1.0212879180908203, | |
| "eval_runtime": 43.2301, | |
| "eval_samples_per_second": 3.585, | |
| "eval_steps_per_second": 0.463, | |
| "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": 1.4620505633437778e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |