| { | |
| "best_metric": 0.9066666666666666, | |
| "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-210", | |
| "epoch": 4.180952380952381, | |
| "eval_steps": 500, | |
| "global_step": 210, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.047619047619047616, | |
| "grad_norm": 7.849704265594482, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 2.3976, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.09523809523809523, | |
| "grad_norm": 7.76956033706665, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 2.3357, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.14285714285714285, | |
| "grad_norm": 15.013307571411133, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 2.282, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.19047619047619047, | |
| "grad_norm": 7.262270450592041, | |
| "learning_rate": 4.4973544973544974e-05, | |
| "loss": 2.0368, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.20476190476190476, | |
| "eval_accuracy": 0.29333333333333333, | |
| "eval_loss": 2.0273361206054688, | |
| "eval_runtime": 7.264, | |
| "eval_samples_per_second": 10.325, | |
| "eval_steps_per_second": 1.377, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.0333333333333334, | |
| "grad_norm": 8.3132963180542, | |
| "learning_rate": 4.232804232804233e-05, | |
| "loss": 1.947, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.0809523809523809, | |
| "grad_norm": 14.42203426361084, | |
| "learning_rate": 3.968253968253968e-05, | |
| "loss": 1.594, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.1285714285714286, | |
| "grad_norm": 11.382922172546387, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 1.2509, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.1761904761904762, | |
| "grad_norm": 9.775425910949707, | |
| "learning_rate": 3.439153439153439e-05, | |
| "loss": 1.1679, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.2047619047619047, | |
| "eval_accuracy": 0.56, | |
| "eval_loss": 1.0918534994125366, | |
| "eval_runtime": 7.1505, | |
| "eval_samples_per_second": 10.489, | |
| "eval_steps_per_second": 1.398, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 2.019047619047619, | |
| "grad_norm": 16.352764129638672, | |
| "learning_rate": 3.1746031746031745e-05, | |
| "loss": 1.0361, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.066666666666667, | |
| "grad_norm": 5.250826358795166, | |
| "learning_rate": 2.91005291005291e-05, | |
| "loss": 0.7379, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.1142857142857143, | |
| "grad_norm": 11.715250968933105, | |
| "learning_rate": 2.6455026455026456e-05, | |
| "loss": 0.7289, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.1619047619047618, | |
| "grad_norm": 16.230215072631836, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 0.5097, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.204761904761905, | |
| "eval_accuracy": 0.8266666666666667, | |
| "eval_loss": 0.5804122090339661, | |
| "eval_runtime": 6.6096, | |
| "eval_samples_per_second": 11.347, | |
| "eval_steps_per_second": 1.513, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 3.0047619047619047, | |
| "grad_norm": 18.1605224609375, | |
| "learning_rate": 2.1164021164021164e-05, | |
| "loss": 0.4609, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.052380952380952, | |
| "grad_norm": 7.160711288452148, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.3043, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 14.182634353637695, | |
| "learning_rate": 1.5873015873015872e-05, | |
| "loss": 0.336, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.1476190476190475, | |
| "grad_norm": 2.916877031326294, | |
| "learning_rate": 1.3227513227513228e-05, | |
| "loss": 0.2992, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.1952380952380954, | |
| "grad_norm": 8.268482208251953, | |
| "learning_rate": 1.0582010582010582e-05, | |
| "loss": 0.2293, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.204761904761905, | |
| "eval_accuracy": 0.8266666666666667, | |
| "eval_loss": 0.4854774475097656, | |
| "eval_runtime": 6.5331, | |
| "eval_samples_per_second": 11.48, | |
| "eval_steps_per_second": 1.531, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 4.038095238095238, | |
| "grad_norm": 7.389148712158203, | |
| "learning_rate": 7.936507936507936e-06, | |
| "loss": 0.3101, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.085714285714285, | |
| "grad_norm": 4.5858259201049805, | |
| "learning_rate": 5.291005291005291e-06, | |
| "loss": 0.1873, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.133333333333334, | |
| "grad_norm": 2.300278425216675, | |
| "learning_rate": 2.6455026455026455e-06, | |
| "loss": 0.2008, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.180952380952381, | |
| "grad_norm": 2.982243537902832, | |
| "learning_rate": 0.0, | |
| "loss": 0.1882, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 4.180952380952381, | |
| "eval_accuracy": 0.9066666666666666, | |
| "eval_loss": 0.3435481786727905, | |
| "eval_runtime": 6.9489, | |
| "eval_samples_per_second": 10.793, | |
| "eval_steps_per_second": 1.439, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 4.180952380952381, | |
| "step": 210, | |
| "total_flos": 2.0736178988651643e+18, | |
| "train_loss": 0.9781298257055737, | |
| "train_runtime": 354.3465, | |
| "train_samples_per_second": 4.741, | |
| "train_steps_per_second": 0.593 | |
| }, | |
| { | |
| "epoch": 4.180952380952381, | |
| "eval_accuracy": 0.8395061728395061, | |
| "eval_loss": 0.3850219249725342, | |
| "eval_runtime": 60.1785, | |
| "eval_samples_per_second": 2.692, | |
| "eval_steps_per_second": 0.349, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 4.180952380952381, | |
| "eval_accuracy": 0.8395061728395061, | |
| "eval_loss": 0.38502195477485657, | |
| "eval_runtime": 14.6842, | |
| "eval_samples_per_second": 11.032, | |
| "eval_steps_per_second": 1.43, | |
| "step": 210 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 210, | |
| "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": 2.0736178988651643e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |