| { | |
| "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": 6.154857635498047, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 2.326, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.13513513513513514, | |
| "grad_norm": 7.193238735198975, | |
| "learning_rate": 4.81203007518797e-05, | |
| "loss": 2.2205, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.20270270270270271, | |
| "grad_norm": 7.481975555419922, | |
| "learning_rate": 4.43609022556391e-05, | |
| "loss": 2.16, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.25675675675675674, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.7993043661117554, | |
| "eval_runtime": 6.6774, | |
| "eval_samples_per_second": 10.483, | |
| "eval_steps_per_second": 1.348, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.0135135135135136, | |
| "grad_norm": 10.227862358093262, | |
| "learning_rate": 4.0601503759398494e-05, | |
| "loss": 1.9946, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.0810810810810811, | |
| "grad_norm": 10.306306838989258, | |
| "learning_rate": 3.6842105263157895e-05, | |
| "loss": 1.5987, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.1486486486486487, | |
| "grad_norm": 10.154570579528809, | |
| "learning_rate": 3.3082706766917295e-05, | |
| "loss": 1.197, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.2162162162162162, | |
| "grad_norm": 8.003480911254883, | |
| "learning_rate": 2.9323308270676693e-05, | |
| "loss": 0.8394, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.2567567567567568, | |
| "eval_accuracy": 0.7, | |
| "eval_loss": 0.8820496201515198, | |
| "eval_runtime": 6.7555, | |
| "eval_samples_per_second": 10.362, | |
| "eval_steps_per_second": 1.332, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.027027027027027, | |
| "grad_norm": 6.113378047943115, | |
| "learning_rate": 2.556390977443609e-05, | |
| "loss": 0.8036, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.0945945945945947, | |
| "grad_norm": 13.039664268493652, | |
| "learning_rate": 2.1804511278195487e-05, | |
| "loss": 0.5384, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.1621621621621623, | |
| "grad_norm": 14.060022354125977, | |
| "learning_rate": 1.8045112781954888e-05, | |
| "loss": 0.6038, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.22972972972973, | |
| "grad_norm": 6.143197536468506, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 0.4228, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.2567567567567566, | |
| "eval_accuracy": 0.8428571428571429, | |
| "eval_loss": 0.5060061812400818, | |
| "eval_runtime": 7.8853, | |
| "eval_samples_per_second": 8.877, | |
| "eval_steps_per_second": 1.141, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 3.0405405405405403, | |
| "grad_norm": 5.005883693695068, | |
| "learning_rate": 1.0526315789473684e-05, | |
| "loss": 0.3779, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.108108108108108, | |
| "grad_norm": 4.216254711151123, | |
| "learning_rate": 6.766917293233083e-06, | |
| "loss": 0.2576, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.175675675675676, | |
| "grad_norm": 8.001497268676758, | |
| "learning_rate": 3.007518796992481e-06, | |
| "loss": 0.3029, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "eval_accuracy": 0.8857142857142857, | |
| "eval_loss": 0.38448596000671387, | |
| "eval_runtime": 7.6412, | |
| "eval_samples_per_second": 9.161, | |
| "eval_steps_per_second": 1.178, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "step": 148, | |
| "total_flos": 1.460491890402263e+18, | |
| "train_loss": 1.0705470426662549, | |
| "train_runtime": 391.8273, | |
| "train_samples_per_second": 3.022, | |
| "train_steps_per_second": 0.378 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "eval_accuracy": 0.8580645161290322, | |
| "eval_loss": 0.4352799952030182, | |
| "eval_runtime": 17.7551, | |
| "eval_samples_per_second": 8.73, | |
| "eval_steps_per_second": 1.126, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 3.22972972972973, | |
| "eval_accuracy": 0.8580645161290322, | |
| "eval_loss": 0.435279905796051, | |
| "eval_runtime": 14.7921, | |
| "eval_samples_per_second": 10.479, | |
| "eval_steps_per_second": 1.352, | |
| "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 | |
| } | |