| { | |
| "best_metric": 0.5714285714285714, | |
| "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-75", | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 9.74156379699707, | |
| "learning_rate": 4.850746268656717e-05, | |
| "loss": 2.3609, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 9.950693130493164, | |
| "learning_rate": 4.104477611940299e-05, | |
| "loss": 2.2224, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 9.031251907348633, | |
| "learning_rate": 3.358208955223881e-05, | |
| "loss": 2.1254, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 9.428235054016113, | |
| "learning_rate": 2.6119402985074626e-05, | |
| "loss": 1.8965, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 16.60507583618164, | |
| "learning_rate": 1.865671641791045e-05, | |
| "loss": 1.9915, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 10.332749366760254, | |
| "learning_rate": 1.119402985074627e-05, | |
| "loss": 1.7514, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 13.279064178466797, | |
| "learning_rate": 3.7313432835820893e-06, | |
| "loss": 1.8974, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5714285714285714, | |
| "eval_loss": 1.5608302354812622, | |
| "eval_runtime": 12.4656, | |
| "eval_samples_per_second": 5.615, | |
| "eval_steps_per_second": 1.444, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 75, | |
| "total_flos": 3.73846047031296e+17, | |
| "train_loss": 2.0024278195699057, | |
| "train_runtime": 118.6782, | |
| "train_samples_per_second": 2.528, | |
| "train_steps_per_second": 0.632 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.3225806451612903, | |
| "eval_loss": 1.8583416938781738, | |
| "eval_runtime": 27.5747, | |
| "eval_samples_per_second": 5.621, | |
| "eval_steps_per_second": 1.414, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.3225806451612903, | |
| "eval_loss": 1.8583415746688843, | |
| "eval_runtime": 28.2276, | |
| "eval_samples_per_second": 5.491, | |
| "eval_steps_per_second": 1.382, | |
| "step": 75 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 75, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 3.73846047031296e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |