| { | |
| "best_metric": 0.9743589743589743, | |
| "best_model_checkpoint": "videomae-base-finetuned-numbers/checkpoint-176", | |
| "epoch": 3.25, | |
| "eval_steps": 500, | |
| "global_step": 176, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.056818181818181816, | |
| "grad_norm": 13.327278137207031, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.8601, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.11363636363636363, | |
| "grad_norm": 18.073780059814453, | |
| "learning_rate": 4.936708860759494e-05, | |
| "loss": 0.7637, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.17045454545454544, | |
| "grad_norm": 21.950044631958008, | |
| "learning_rate": 4.6202531645569625e-05, | |
| "loss": 0.8635, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.22727272727272727, | |
| "grad_norm": 11.889289855957031, | |
| "learning_rate": 4.3037974683544305e-05, | |
| "loss": 0.7592, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_accuracy": 0.8461538461538461, | |
| "eval_f1": 0.8479076479076479, | |
| "eval_loss": 0.6377912759780884, | |
| "eval_precision": 0.8757575757575757, | |
| "eval_recall": 0.8560606060606059, | |
| "eval_runtime": 77.2345, | |
| "eval_samples_per_second": 0.505, | |
| "eval_steps_per_second": 0.129, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.0340909090909092, | |
| "grad_norm": 6.685118198394775, | |
| "learning_rate": 3.987341772151899e-05, | |
| "loss": 0.6971, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 5.797889232635498, | |
| "learning_rate": 3.670886075949367e-05, | |
| "loss": 0.4348, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.1477272727272727, | |
| "grad_norm": 2.289470672607422, | |
| "learning_rate": 3.354430379746836e-05, | |
| "loss": 0.4708, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.2045454545454546, | |
| "grad_norm": 1.6386131048202515, | |
| "learning_rate": 3.0379746835443042e-05, | |
| "loss": 0.296, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_accuracy": 0.8974358974358975, | |
| "eval_f1": 0.8805194805194806, | |
| "eval_loss": 0.3026905655860901, | |
| "eval_precision": 0.9090909090909091, | |
| "eval_recall": 0.8863636363636362, | |
| "eval_runtime": 76.9133, | |
| "eval_samples_per_second": 0.507, | |
| "eval_steps_per_second": 0.13, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 2.0113636363636362, | |
| "grad_norm": 0.7323519587516785, | |
| "learning_rate": 2.7215189873417722e-05, | |
| "loss": 0.4399, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.0681818181818183, | |
| "grad_norm": 7.895605564117432, | |
| "learning_rate": 2.4050632911392405e-05, | |
| "loss": 0.2948, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.125, | |
| "grad_norm": 27.428850173950195, | |
| "learning_rate": 2.088607594936709e-05, | |
| "loss": 0.3388, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "grad_norm": 0.781851053237915, | |
| "learning_rate": 1.7721518987341772e-05, | |
| "loss": 0.2823, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.2386363636363638, | |
| "grad_norm": 22.52676773071289, | |
| "learning_rate": 1.4556962025316457e-05, | |
| "loss": 0.2144, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "eval_accuracy": 0.9487179487179487, | |
| "eval_f1": 0.9376623376623378, | |
| "eval_loss": 0.12886768579483032, | |
| "eval_precision": 0.9545454545454546, | |
| "eval_recall": 0.9393939393939393, | |
| "eval_runtime": 65.5077, | |
| "eval_samples_per_second": 0.595, | |
| "eval_steps_per_second": 0.153, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 3.0454545454545454, | |
| "grad_norm": 16.04294204711914, | |
| "learning_rate": 1.139240506329114e-05, | |
| "loss": 0.2912, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.102272727272727, | |
| "grad_norm": 0.34404367208480835, | |
| "learning_rate": 8.227848101265822e-06, | |
| "loss": 0.0882, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.159090909090909, | |
| "grad_norm": 0.7014194130897522, | |
| "learning_rate": 5.063291139240506e-06, | |
| "loss": 0.083, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.215909090909091, | |
| "grad_norm": 22.440013885498047, | |
| "learning_rate": 1.8987341772151901e-06, | |
| "loss": 0.1331, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "eval_accuracy": 0.9743589743589743, | |
| "eval_f1": 0.9688311688311688, | |
| "eval_loss": 0.09583019465208054, | |
| "eval_precision": 0.9772727272727273, | |
| "eval_recall": 0.9696969696969696, | |
| "eval_runtime": 66.5184, | |
| "eval_samples_per_second": 0.586, | |
| "eval_steps_per_second": 0.15, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "step": 176, | |
| "total_flos": 8.922538555213087e+17, | |
| "train_loss": 0.41998768225312233, | |
| "train_runtime": 1661.3641, | |
| "train_samples_per_second": 0.424, | |
| "train_steps_per_second": 0.106 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "eval_accuracy": 0.8222222222222222, | |
| "eval_f1": 0.8014692378328742, | |
| "eval_loss": 0.3432573676109314, | |
| "eval_precision": 0.8761904761904762, | |
| "eval_recall": 0.8181818181818182, | |
| "eval_runtime": 74.9292, | |
| "eval_samples_per_second": 0.601, | |
| "eval_steps_per_second": 0.16, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "eval_accuracy": 0.8222222222222222, | |
| "eval_f1": 0.8014692378328742, | |
| "eval_loss": 0.3432573974132538, | |
| "eval_precision": 0.8761904761904762, | |
| "eval_recall": 0.8181818181818182, | |
| "eval_runtime": 76.5031, | |
| "eval_samples_per_second": 0.588, | |
| "eval_steps_per_second": 0.157, | |
| "step": 176 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 176, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 8.922538555213087e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |