| { |
| "best_metric": 0.915625, |
| "best_model_checkpoint": "vit-base-patch16-224-finalterm/checkpoint-140", |
| "epoch": 15.0, |
| "eval_steps": 500, |
| "global_step": 150, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 3.922755241394043, |
| "learning_rate": 3.3333333333333335e-05, |
| "loss": 1.4514, |
| "step": 10 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.559375, |
| "eval_loss": 1.167618989944458, |
| "eval_runtime": 3.1299, |
| "eval_samples_per_second": 102.239, |
| "eval_steps_per_second": 3.195, |
| "step": 10 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 3.1222116947174072, |
| "learning_rate": 4.814814814814815e-05, |
| "loss": 1.0123, |
| "step": 20 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.7, |
| "eval_loss": 0.8034278750419617, |
| "eval_runtime": 3.1134, |
| "eval_samples_per_second": 102.781, |
| "eval_steps_per_second": 3.212, |
| "step": 20 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 2.397467613220215, |
| "learning_rate": 4.4444444444444447e-05, |
| "loss": 0.6949, |
| "step": 30 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.828125, |
| "eval_loss": 0.5371649861335754, |
| "eval_runtime": 3.1115, |
| "eval_samples_per_second": 102.843, |
| "eval_steps_per_second": 3.214, |
| "step": 30 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 2.74021315574646, |
| "learning_rate": 4.074074074074074e-05, |
| "loss": 0.5385, |
| "step": 40 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.871875, |
| "eval_loss": 0.4333480894565582, |
| "eval_runtime": 3.1377, |
| "eval_samples_per_second": 101.986, |
| "eval_steps_per_second": 3.187, |
| "step": 40 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 3.3417820930480957, |
| "learning_rate": 3.7037037037037037e-05, |
| "loss": 0.4984, |
| "step": 50 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.884375, |
| "eval_loss": 0.3638252317905426, |
| "eval_runtime": 3.1292, |
| "eval_samples_per_second": 102.261, |
| "eval_steps_per_second": 3.196, |
| "step": 50 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 2.3100404739379883, |
| "learning_rate": 3.3333333333333335e-05, |
| "loss": 0.4382, |
| "step": 60 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.89375, |
| "eval_loss": 0.31991782784461975, |
| "eval_runtime": 3.1525, |
| "eval_samples_per_second": 101.508, |
| "eval_steps_per_second": 3.172, |
| "step": 60 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 2.219219207763672, |
| "learning_rate": 2.962962962962963e-05, |
| "loss": 0.3739, |
| "step": 70 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.9, |
| "eval_loss": 0.3272128105163574, |
| "eval_runtime": 3.1159, |
| "eval_samples_per_second": 102.7, |
| "eval_steps_per_second": 3.209, |
| "step": 70 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 2.4531455039978027, |
| "learning_rate": 2.5925925925925925e-05, |
| "loss": 0.3692, |
| "step": 80 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.890625, |
| "eval_loss": 0.3230211138725281, |
| "eval_runtime": 3.0903, |
| "eval_samples_per_second": 103.548, |
| "eval_steps_per_second": 3.236, |
| "step": 80 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 2.368349075317383, |
| "learning_rate": 2.2222222222222223e-05, |
| "loss": 0.3705, |
| "step": 90 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.896875, |
| "eval_loss": 0.33077651262283325, |
| "eval_runtime": 3.1039, |
| "eval_samples_per_second": 103.096, |
| "eval_steps_per_second": 3.222, |
| "step": 90 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 1.733303427696228, |
| "learning_rate": 1.8518518518518518e-05, |
| "loss": 0.3458, |
| "step": 100 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_accuracy": 0.9, |
| "eval_loss": 0.29199448227882385, |
| "eval_runtime": 3.1065, |
| "eval_samples_per_second": 103.009, |
| "eval_steps_per_second": 3.219, |
| "step": 100 |
| }, |
| { |
| "epoch": 11.0, |
| "grad_norm": 2.134023904800415, |
| "learning_rate": 1.4814814814814815e-05, |
| "loss": 0.3176, |
| "step": 110 |
| }, |
| { |
| "epoch": 11.0, |
| "eval_accuracy": 0.90625, |
| "eval_loss": 0.28801438212394714, |
| "eval_runtime": 3.1103, |
| "eval_samples_per_second": 102.885, |
| "eval_steps_per_second": 3.215, |
| "step": 110 |
| }, |
| { |
| "epoch": 12.0, |
| "grad_norm": 2.6795620918273926, |
| "learning_rate": 1.1111111111111112e-05, |
| "loss": 0.3101, |
| "step": 120 |
| }, |
| { |
| "epoch": 12.0, |
| "eval_accuracy": 0.896875, |
| "eval_loss": 0.3041202425956726, |
| "eval_runtime": 3.1014, |
| "eval_samples_per_second": 103.178, |
| "eval_steps_per_second": 3.224, |
| "step": 120 |
| }, |
| { |
| "epoch": 13.0, |
| "grad_norm": 2.048682451248169, |
| "learning_rate": 7.4074074074074075e-06, |
| "loss": 0.3224, |
| "step": 130 |
| }, |
| { |
| "epoch": 13.0, |
| "eval_accuracy": 0.903125, |
| "eval_loss": 0.27769678831100464, |
| "eval_runtime": 3.0997, |
| "eval_samples_per_second": 103.235, |
| "eval_steps_per_second": 3.226, |
| "step": 130 |
| }, |
| { |
| "epoch": 14.0, |
| "grad_norm": 1.949576735496521, |
| "learning_rate": 3.7037037037037037e-06, |
| "loss": 0.2946, |
| "step": 140 |
| }, |
| { |
| "epoch": 14.0, |
| "eval_accuracy": 0.915625, |
| "eval_loss": 0.26764923334121704, |
| "eval_runtime": 3.1214, |
| "eval_samples_per_second": 102.519, |
| "eval_steps_per_second": 3.204, |
| "step": 140 |
| }, |
| { |
| "epoch": 15.0, |
| "grad_norm": 2.0188732147216797, |
| "learning_rate": 0.0, |
| "loss": 0.2693, |
| "step": 150 |
| }, |
| { |
| "epoch": 15.0, |
| "eval_accuracy": 0.9125, |
| "eval_loss": 0.27088016271591187, |
| "eval_runtime": 3.1019, |
| "eval_samples_per_second": 103.162, |
| "eval_steps_per_second": 3.224, |
| "step": 150 |
| }, |
| { |
| "epoch": 15.0, |
| "step": 150, |
| "total_flos": 1.4878728707899392e+18, |
| "train_loss": 0.5071320374806721, |
| "train_runtime": 447.0543, |
| "train_samples_per_second": 42.948, |
| "train_steps_per_second": 0.336 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 150, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 15, |
| "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": 1.4878728707899392e+18, |
| "train_batch_size": 32, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
