| { |
| "best_metric": 0.9045333333333333, |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0134/checkpoints/checkpoint-2997", |
| "epoch": 9.0, |
| "eval_steps": 500, |
| "global_step": 2997, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 31.496679306030273, |
| "learning_rate": 0.0004849231551964771, |
| "loss": 1.3497, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.8274666666666667, |
| "eval_loss": 0.6115483641624451, |
| "eval_runtime": 8.28, |
| "eval_samples_per_second": 452.897, |
| "eval_steps_per_second": 7.126, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 28.855010986328125, |
| "learning_rate": 0.0004415111107797445, |
| "loss": 0.4459, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8573333333333333, |
| "eval_loss": 0.5151609182357788, |
| "eval_runtime": 8.4555, |
| "eval_samples_per_second": 443.5, |
| "eval_steps_per_second": 6.978, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 16.289453506469727, |
| "learning_rate": 0.000375, |
| "loss": 0.2469, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8786666666666667, |
| "eval_loss": 0.443855345249176, |
| "eval_runtime": 8.2648, |
| "eval_samples_per_second": 453.732, |
| "eval_steps_per_second": 7.139, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 23.52910804748535, |
| "learning_rate": 0.00029341204441673266, |
| "loss": 0.1483, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8922666666666667, |
| "eval_loss": 0.43019700050354004, |
| "eval_runtime": 8.2468, |
| "eval_samples_per_second": 454.72, |
| "eval_steps_per_second": 7.154, |
| "step": 1332 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 35.86688232421875, |
| "learning_rate": 0.00020658795558326743, |
| "loss": 0.0884, |
| "step": 1665 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.8864, |
| "eval_loss": 0.45890143513679504, |
| "eval_runtime": 8.2732, |
| "eval_samples_per_second": 453.27, |
| "eval_steps_per_second": 7.131, |
| "step": 1665 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 35.50931930541992, |
| "learning_rate": 0.00012500000000000006, |
| "loss": 0.0476, |
| "step": 1998 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.8936, |
| "eval_loss": 0.4689825475215912, |
| "eval_runtime": 8.255, |
| "eval_samples_per_second": 454.268, |
| "eval_steps_per_second": 7.147, |
| "step": 1998 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 27.811569213867188, |
| "learning_rate": 5.848888922025553e-05, |
| "loss": 0.0305, |
| "step": 2331 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.9010666666666667, |
| "eval_loss": 0.43023476004600525, |
| "eval_runtime": 10.5207, |
| "eval_samples_per_second": 356.439, |
| "eval_steps_per_second": 5.608, |
| "step": 2331 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 39.86514663696289, |
| "learning_rate": 1.5076844803522921e-05, |
| "loss": 0.0218, |
| "step": 2664 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.9010666666666667, |
| "eval_loss": 0.43928220868110657, |
| "eval_runtime": 8.0458, |
| "eval_samples_per_second": 466.08, |
| "eval_steps_per_second": 7.333, |
| "step": 2664 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 22.864826202392578, |
| "learning_rate": 0.0, |
| "loss": 0.0172, |
| "step": 2997 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.9045333333333333, |
| "eval_loss": 0.4326859414577484, |
| "eval_runtime": 8.1446, |
| "eval_samples_per_second": 460.426, |
| "eval_steps_per_second": 7.244, |
| "step": 2997 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 2997, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 9, |
| "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": 7.3587863155968e+18, |
| "train_batch_size": 64, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
