| { |
| "best_metric": 0.7856, |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0579/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": 278.3972473144531, |
| "learning_rate": 2.6666666666666667e-05, |
| "loss": 3.8546, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.288, |
| "eval_loss": 3.737907886505127, |
| "eval_runtime": 7.5106, |
| "eval_samples_per_second": 499.293, |
| "eval_steps_per_second": 7.856, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 96.56761169433594, |
| "learning_rate": 2.3333333333333336e-05, |
| "loss": 3.3529, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.5237333333333334, |
| "eval_loss": 2.898594617843628, |
| "eval_runtime": 7.9141, |
| "eval_samples_per_second": 473.838, |
| "eval_steps_per_second": 7.455, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 168.0374298095703, |
| "learning_rate": 1.9999999999999998e-05, |
| "loss": 2.186, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.6512, |
| "eval_loss": 1.6011171340942383, |
| "eval_runtime": 7.5188, |
| "eval_samples_per_second": 498.748, |
| "eval_steps_per_second": 7.847, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 359.4327087402344, |
| "learning_rate": 1.6666666666666667e-05, |
| "loss": 1.4029, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.72, |
| "eval_loss": 1.171289324760437, |
| "eval_runtime": 7.7576, |
| "eval_samples_per_second": 483.398, |
| "eval_steps_per_second": 7.605, |
| "step": 1332 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 287.1774597167969, |
| "learning_rate": 1.3333333333333333e-05, |
| "loss": 1.0813, |
| "step": 1665 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.7458666666666667, |
| "eval_loss": 0.9461485743522644, |
| "eval_runtime": 7.4511, |
| "eval_samples_per_second": 503.285, |
| "eval_steps_per_second": 7.918, |
| "step": 1665 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 205.04466247558594, |
| "learning_rate": 9.999999999999999e-06, |
| "loss": 0.9285, |
| "step": 1998 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.7568, |
| "eval_loss": 0.8765926957130432, |
| "eval_runtime": 7.9345, |
| "eval_samples_per_second": 472.618, |
| "eval_steps_per_second": 7.436, |
| "step": 1998 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 309.07275390625, |
| "learning_rate": 6.666666666666667e-06, |
| "loss": 0.8419, |
| "step": 2331 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.7762666666666667, |
| "eval_loss": 0.7876874804496765, |
| "eval_runtime": 7.4785, |
| "eval_samples_per_second": 501.438, |
| "eval_steps_per_second": 7.889, |
| "step": 2331 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 120.86518859863281, |
| "learning_rate": 3.3333333333333333e-06, |
| "loss": 0.7968, |
| "step": 2664 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.7794666666666666, |
| "eval_loss": 0.760668158531189, |
| "eval_runtime": 7.5142, |
| "eval_samples_per_second": 499.052, |
| "eval_steps_per_second": 7.852, |
| "step": 2664 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 151.1305389404297, |
| "learning_rate": 0.0, |
| "loss": 0.7723, |
| "step": 2997 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.7856, |
| "eval_loss": 0.7535416483879089, |
| "eval_runtime": 7.9769, |
| "eval_samples_per_second": 470.11, |
| "eval_steps_per_second": 7.396, |
| "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 |
| } |
|
|
