{ "best_metric": 0.9219512195121952, "best_model_checkpoint": "videomae-base-finetuned-kinetics-finetuned-lipsync-subset-1/checkpoint-156", "epoch": 3.25, "eval_steps": 500, "global_step": 208, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.05, "learning_rate": 2.380952380952381e-05, "loss": 0.7283, "step": 10 }, { "epoch": 0.1, "learning_rate": 4.761904761904762e-05, "loss": 0.6955, "step": 20 }, { "epoch": 0.14, "learning_rate": 4.759358288770054e-05, "loss": 0.6767, "step": 30 }, { "epoch": 0.19, "learning_rate": 4.491978609625669e-05, "loss": 0.6488, "step": 40 }, { "epoch": 0.24, "learning_rate": 4.224598930481284e-05, "loss": 0.6552, "step": 50 }, { "epoch": 0.25, "eval_accuracy": 0.5609756097560976, "eval_loss": 0.648867666721344, "eval_runtime": 136.2147, "eval_samples_per_second": 1.505, "eval_steps_per_second": 0.191, "step": 52 }, { "epoch": 1.04, "learning_rate": 3.957219251336899e-05, "loss": 0.514, "step": 60 }, { "epoch": 1.09, "learning_rate": 3.6898395721925136e-05, "loss": 0.3894, "step": 70 }, { "epoch": 1.13, "learning_rate": 3.4224598930481284e-05, "loss": 0.4006, "step": 80 }, { "epoch": 1.18, "learning_rate": 3.155080213903743e-05, "loss": 0.3072, "step": 90 }, { "epoch": 1.23, "learning_rate": 2.8877005347593582e-05, "loss": 0.4105, "step": 100 }, { "epoch": 1.25, "eval_accuracy": 0.8780487804878049, "eval_loss": 0.3532041609287262, "eval_runtime": 131.8033, "eval_samples_per_second": 1.555, "eval_steps_per_second": 0.197, "step": 104 }, { "epoch": 2.03, "learning_rate": 2.6203208556149733e-05, "loss": 0.2976, "step": 110 }, { "epoch": 2.08, "learning_rate": 2.3529411764705884e-05, "loss": 0.2505, "step": 120 }, { "epoch": 2.12, "learning_rate": 2.0855614973262035e-05, "loss": 0.1904, "step": 130 }, { "epoch": 2.17, "learning_rate": 1.8181818181818182e-05, "loss": 0.1751, "step": 140 }, { "epoch": 2.22, "learning_rate": 1.5508021390374333e-05, "loss": 0.5187, "step": 150 }, { "epoch": 2.25, "eval_accuracy": 0.9219512195121952, "eval_loss": 0.2863960862159729, "eval_runtime": 102.7728, "eval_samples_per_second": 1.995, "eval_steps_per_second": 0.253, "step": 156 }, { "epoch": 3.02, "learning_rate": 1.2834224598930484e-05, "loss": 0.1526, "step": 160 }, { "epoch": 3.07, "learning_rate": 1.0160427807486631e-05, "loss": 0.188, "step": 170 }, { "epoch": 3.12, "learning_rate": 7.4866310160427806e-06, "loss": 0.1734, "step": 180 }, { "epoch": 3.16, "learning_rate": 4.812834224598931e-06, "loss": 0.178, "step": 190 }, { "epoch": 3.21, "learning_rate": 2.1390374331550802e-06, "loss": 0.2265, "step": 200 }, { "epoch": 3.25, "eval_accuracy": 0.9121951219512195, "eval_loss": 0.30466458201408386, "eval_runtime": 104.4316, "eval_samples_per_second": 1.963, "eval_steps_per_second": 0.249, "step": 208 }, { "epoch": 3.25, "step": 208, "total_flos": 2.073451476880982e+18, "train_loss": 0.38049622854361165, "train_runtime": 1977.8449, "train_samples_per_second": 0.841, "train_steps_per_second": 0.105 }, { "epoch": 3.25, "eval_accuracy": 0.8106796116504854, "eval_loss": 0.5757558345794678, "eval_runtime": 152.3675, "eval_samples_per_second": 1.352, "eval_steps_per_second": 0.171, "step": 208 }, { "epoch": 3.25, "eval_accuracy": 0.8106796116504854, "eval_loss": 0.5757559537887573, "eval_runtime": 154.6888, "eval_samples_per_second": 1.332, "eval_steps_per_second": 0.168, "step": 208 } ], "logging_steps": 10, "max_steps": 208, "num_train_epochs": 9223372036854775807, "save_steps": 500, "total_flos": 2.073451476880982e+18, "trial_name": null, "trial_params": null }