{ "best_metric": 1.0, "best_model_checkpoint": "videomae-base-OKN-Test-V15\\checkpoint-92", "epoch": 4.2, "eval_steps": 500, "global_step": 230, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.04, "learning_rate": 2.173913043478261e-05, "loss": 0.6817, "step": 10 }, { "epoch": 0.09, "learning_rate": 4.347826086956522e-05, "loss": 0.447, "step": 20 }, { "epoch": 0.13, "learning_rate": 4.830917874396135e-05, "loss": 0.4195, "step": 30 }, { "epoch": 0.17, "learning_rate": 4.589371980676328e-05, "loss": 0.465, "step": 40 }, { "epoch": 0.2, "eval_accuracy": 0.9047619047619048, "eval_loss": 0.1457657516002655, "eval_runtime": 22.0681, "eval_samples_per_second": 0.952, "eval_steps_per_second": 0.272, "step": 46 }, { "epoch": 1.02, "learning_rate": 4.347826086956522e-05, "loss": 0.2322, "step": 50 }, { "epoch": 1.06, "learning_rate": 4.106280193236715e-05, "loss": 0.0717, "step": 60 }, { "epoch": 1.1, "learning_rate": 3.864734299516908e-05, "loss": 0.0103, "step": 70 }, { "epoch": 1.15, "learning_rate": 3.6231884057971014e-05, "loss": 0.001, "step": 80 }, { "epoch": 1.19, "learning_rate": 3.381642512077295e-05, "loss": 0.0115, "step": 90 }, { "epoch": 1.2, "eval_accuracy": 1.0, "eval_loss": 0.00043274767813272774, "eval_runtime": 24.9569, "eval_samples_per_second": 0.841, "eval_steps_per_second": 0.24, "step": 92 }, { "epoch": 2.03, "learning_rate": 3.140096618357488e-05, "loss": 0.0004, "step": 100 }, { "epoch": 2.08, "learning_rate": 2.8985507246376814e-05, "loss": 0.0004, "step": 110 }, { "epoch": 2.12, "learning_rate": 2.6570048309178748e-05, "loss": 0.0025, "step": 120 }, { "epoch": 2.17, "learning_rate": 2.4154589371980676e-05, "loss": 0.17, "step": 130 }, { "epoch": 2.2, "eval_accuracy": 0.9523809523809523, "eval_loss": 0.09127208590507507, "eval_runtime": 23.3382, "eval_samples_per_second": 0.9, "eval_steps_per_second": 0.257, "step": 138 }, { "epoch": 3.01, "learning_rate": 2.173913043478261e-05, "loss": 0.0005, "step": 140 }, { "epoch": 3.05, "learning_rate": 1.932367149758454e-05, "loss": 0.0004, "step": 150 }, { "epoch": 3.1, "learning_rate": 1.6908212560386476e-05, "loss": 0.0004, "step": 160 }, { "epoch": 3.14, "learning_rate": 1.4492753623188407e-05, "loss": 0.0003, "step": 170 }, { "epoch": 3.18, "learning_rate": 1.2077294685990338e-05, "loss": 0.0652, "step": 180 }, { "epoch": 3.2, "eval_accuracy": 1.0, "eval_loss": 0.0002443942066747695, "eval_runtime": 22.9752, "eval_samples_per_second": 0.914, "eval_steps_per_second": 0.261, "step": 184 }, { "epoch": 4.03, "learning_rate": 9.66183574879227e-06, "loss": 0.0003, "step": 190 }, { "epoch": 4.07, "learning_rate": 7.246376811594203e-06, "loss": 0.0003, "step": 200 }, { "epoch": 4.11, "learning_rate": 4.830917874396135e-06, "loss": 0.0002, "step": 210 }, { "epoch": 4.16, "learning_rate": 2.4154589371980677e-06, "loss": 0.0002, "step": 220 }, { "epoch": 4.2, "learning_rate": 0.0, "loss": 0.0002, "step": 230 }, { "epoch": 4.2, "eval_accuracy": 1.0, "eval_loss": 0.00022312559303827584, "eval_runtime": 24.2112, "eval_samples_per_second": 0.867, "eval_steps_per_second": 0.248, "step": 230 }, { "epoch": 4.2, "step": 230, "total_flos": 1.1463794223140045e+18, "train_loss": 0.11223210689166317, "train_runtime": 2573.7407, "train_samples_per_second": 0.357, "train_steps_per_second": 0.089 }, { "epoch": 4.2, "eval_accuracy": 1.0, "eval_loss": 0.00043274767813272774, "eval_runtime": 23.6483, "eval_samples_per_second": 0.888, "eval_steps_per_second": 0.254, "step": 230 }, { "epoch": 4.2, "eval_accuracy": 1.0, "eval_loss": 0.00044166212319396436, "eval_runtime": 23.3044, "eval_samples_per_second": 0.901, "eval_steps_per_second": 0.257, "step": 230 } ], "logging_steps": 10, "max_steps": 230, "num_train_epochs": 9223372036854775807, "save_steps": 500, "total_flos": 1.1463794223140045e+18, "trial_name": null, "trial_params": null }