{ "best_metric": 1.0, "best_model_checkpoint": "videomae-base-finetuned-kinetics-finetuned-ucf101-subset/checkpoint-19", "epoch": 7.101351351351352, "eval_steps": 500, "global_step": 148, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.07, "learning_rate": 3.3333333333333335e-05, "loss": 2.2587, "step": 10 }, { "epoch": 0.13, "eval_accuracy": 1.0, "eval_loss": 1.2644267082214355, "eval_runtime": 49.3463, "eval_samples_per_second": 1.419, "eval_steps_per_second": 0.101, "step": 19 }, { "epoch": 1.01, "learning_rate": 4.81203007518797e-05, "loss": 1.7115, "step": 20 }, { "epoch": 1.07, "learning_rate": 4.43609022556391e-05, "loss": 0.6711, "step": 30 }, { "epoch": 1.13, "eval_accuracy": 1.0, "eval_loss": 0.20978431403636932, "eval_runtime": 49.4291, "eval_samples_per_second": 1.416, "eval_steps_per_second": 0.101, "step": 38 }, { "epoch": 2.01, "learning_rate": 4.0601503759398494e-05, "loss": 0.2568, "step": 40 }, { "epoch": 2.08, "learning_rate": 3.6842105263157895e-05, "loss": 0.1355, "step": 50 }, { "epoch": 2.13, "eval_accuracy": 1.0, "eval_loss": 0.04650784656405449, "eval_runtime": 50.0125, "eval_samples_per_second": 1.4, "eval_steps_per_second": 0.1, "step": 57 }, { "epoch": 3.02, "learning_rate": 3.3082706766917295e-05, "loss": 0.0399, "step": 60 }, { "epoch": 3.09, "learning_rate": 2.9323308270676693e-05, "loss": 0.0295, "step": 70 }, { "epoch": 3.13, "eval_accuracy": 0.9857142857142858, "eval_loss": 0.04307792708277702, "eval_runtime": 50.096, "eval_samples_per_second": 1.397, "eval_steps_per_second": 0.1, "step": 76 }, { "epoch": 4.03, "learning_rate": 2.556390977443609e-05, "loss": 0.0552, "step": 80 }, { "epoch": 4.09, "learning_rate": 2.1804511278195487e-05, "loss": 0.0155, "step": 90 }, { "epoch": 4.13, "eval_accuracy": 1.0, "eval_loss": 0.022627854719758034, "eval_runtime": 50.2169, "eval_samples_per_second": 1.394, "eval_steps_per_second": 0.1, "step": 95 }, { "epoch": 5.03, "learning_rate": 1.8045112781954888e-05, "loss": 0.0159, "step": 100 }, { "epoch": 5.1, "learning_rate": 1.4285714285714285e-05, "loss": 0.0175, "step": 110 }, { "epoch": 5.13, "eval_accuracy": 1.0, "eval_loss": 0.017812443897128105, "eval_runtime": 49.4188, "eval_samples_per_second": 1.416, "eval_steps_per_second": 0.101, "step": 114 }, { "epoch": 6.04, "learning_rate": 1.0526315789473684e-05, "loss": 0.011, "step": 120 }, { "epoch": 6.11, "learning_rate": 6.766917293233083e-06, "loss": 0.0168, "step": 130 }, { "epoch": 6.13, "eval_accuracy": 1.0, "eval_loss": 0.01800677739083767, "eval_runtime": 49.5979, "eval_samples_per_second": 1.411, "eval_steps_per_second": 0.101, "step": 133 }, { "epoch": 7.05, "learning_rate": 3.007518796992481e-06, "loss": 0.008, "step": 140 }, { "epoch": 7.1, "eval_accuracy": 1.0, "eval_loss": 0.018371539190411568, "eval_runtime": 55.2031, "eval_samples_per_second": 1.268, "eval_steps_per_second": 0.091, "step": 148 }, { "epoch": 7.1, "step": 148, "total_flos": 2.915999166844109e+18, "train_loss": 0.35482698079903385, "train_runtime": 4200.8219, "train_samples_per_second": 0.564, "train_steps_per_second": 0.035 }, { "epoch": 7.1, "eval_accuracy": 0.9806451612903225, "eval_loss": 1.2308675050735474, "eval_runtime": 113.8405, "eval_samples_per_second": 1.362, "eval_steps_per_second": 0.088, "step": 148 }, { "epoch": 7.1, "eval_accuracy": 0.9806451612903225, "eval_loss": 1.2308673858642578, "eval_runtime": 118.6696, "eval_samples_per_second": 1.306, "eval_steps_per_second": 0.084, "step": 148 } ], "logging_steps": 10, "max_steps": 148, "num_input_tokens_seen": 0, "num_train_epochs": 9223372036854775807, "save_steps": 500, "total_flos": 2.915999166844109e+18, "train_batch_size": 16, "trial_name": null, "trial_params": null }