| { | |
| "best_metric": 0.04637409746646881, | |
| "best_model_checkpoint": "./vit-beans_leaves_disease/checkpoint-200", | |
| "epoch": 4.0, | |
| "eval_steps": 100, | |
| "global_step": 260, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15384615384615385, | |
| "grad_norm": 2.1672122478485107, | |
| "learning_rate": 0.00019230769230769233, | |
| "loss": 0.8581, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "grad_norm": 1.494405746459961, | |
| "learning_rate": 0.00018461538461538463, | |
| "loss": 0.3656, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.46153846153846156, | |
| "grad_norm": 2.3154523372650146, | |
| "learning_rate": 0.00017692307692307693, | |
| "loss": 0.2064, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 6.511839866638184, | |
| "learning_rate": 0.00016923076923076923, | |
| "loss": 0.2592, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 0.5855913162231445, | |
| "learning_rate": 0.00016153846153846155, | |
| "loss": 0.1352, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.9230769230769231, | |
| "grad_norm": 7.207978248596191, | |
| "learning_rate": 0.00015384615384615385, | |
| "loss": 0.2, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.0769230769230769, | |
| "grad_norm": 0.31788188219070435, | |
| "learning_rate": 0.00014615384615384615, | |
| "loss": 0.1526, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "grad_norm": 0.18049748241901398, | |
| "learning_rate": 0.00013846153846153847, | |
| "loss": 0.1355, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.3846153846153846, | |
| "grad_norm": 0.1663343608379364, | |
| "learning_rate": 0.00013076923076923077, | |
| "loss": 0.0904, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 4.684115409851074, | |
| "learning_rate": 0.0001230769230769231, | |
| "loss": 0.1672, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "eval_accuracy": 0.9473684210526315, | |
| "eval_loss": 0.18420261144638062, | |
| "eval_runtime": 2.0629, | |
| "eval_samples_per_second": 64.473, | |
| "eval_steps_per_second": 8.241, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.6923076923076923, | |
| "grad_norm": 2.094054698944092, | |
| "learning_rate": 0.00011538461538461538, | |
| "loss": 0.1905, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "grad_norm": 4.573807716369629, | |
| "learning_rate": 0.0001076923076923077, | |
| "loss": 0.1537, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.21059000492095947, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0483, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.1538461538461537, | |
| "grad_norm": 0.11981552094221115, | |
| "learning_rate": 9.230769230769232e-05, | |
| "loss": 0.0287, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 0.6703248620033264, | |
| "learning_rate": 8.461538461538461e-05, | |
| "loss": 0.0344, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "grad_norm": 0.09458092600107193, | |
| "learning_rate": 7.692307692307693e-05, | |
| "loss": 0.0433, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.6153846153846154, | |
| "grad_norm": 0.08353827148675919, | |
| "learning_rate": 6.923076923076924e-05, | |
| "loss": 0.0209, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.769230769230769, | |
| "grad_norm": 0.09006886929273605, | |
| "learning_rate": 6.153846153846155e-05, | |
| "loss": 0.0401, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.9230769230769234, | |
| "grad_norm": 0.07661747187376022, | |
| "learning_rate": 5.384615384615385e-05, | |
| "loss": 0.036, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 0.06653212010860443, | |
| "learning_rate": 4.615384615384616e-05, | |
| "loss": 0.03, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "eval_accuracy": 0.9924812030075187, | |
| "eval_loss": 0.04637409746646881, | |
| "eval_runtime": 1.3925, | |
| "eval_samples_per_second": 95.511, | |
| "eval_steps_per_second": 12.208, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.230769230769231, | |
| "grad_norm": 0.12397517263889313, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 0.0131, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.3846153846153846, | |
| "grad_norm": 0.06832624971866608, | |
| "learning_rate": 3.0769230769230774e-05, | |
| "loss": 0.0143, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.5384615384615383, | |
| "grad_norm": 0.06463640928268433, | |
| "learning_rate": 2.307692307692308e-05, | |
| "loss": 0.0124, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.6923076923076925, | |
| "grad_norm": 0.06494957208633423, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 0.0123, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 9.875919342041016, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.0269, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.06551710516214371, | |
| "learning_rate": 0.0, | |
| "loss": 0.0123, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 260, | |
| "total_flos": 3.205097416476426e+17, | |
| "train_loss": 0.12644467391073705, | |
| "train_runtime": 75.7545, | |
| "train_samples_per_second": 54.597, | |
| "train_steps_per_second": 3.432 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 260, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.205097416476426e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |