| { | |
| "best_metric": 0.021338362246751785, | |
| "best_model_checkpoint": "/workspace/training_output/age-gender-vit-large-patch16-224-in21k-v2/checkpoint-9788", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 12235, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.918267266040049e-05, | |
| "loss": 0.3012, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 1.836534532080098e-05, | |
| "loss": 0.2093, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 1.7548017981201474e-05, | |
| "loss": 0.1836, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 1.6730690641601963e-05, | |
| "loss": 0.1758, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9863925883034164, | |
| "eval_loss": 0.04839900881052017, | |
| "eval_runtime": 252.348, | |
| "eval_samples_per_second": 13.687, | |
| "eval_steps_per_second": 1.712, | |
| "step": 2447 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 1.5913363302002453e-05, | |
| "loss": 0.157, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 1.5096035962402944e-05, | |
| "loss": 0.1331, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 1.4278708622803434e-05, | |
| "loss": 0.1351, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 1.3461381283203925e-05, | |
| "loss": 0.1333, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 1.2644053943604415e-05, | |
| "loss": 0.1123, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.980023161551824, | |
| "eval_loss": 0.0850570872426033, | |
| "eval_runtime": 231.2388, | |
| "eval_samples_per_second": 14.937, | |
| "eval_steps_per_second": 1.868, | |
| "step": 4894 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 1.1826726604004906e-05, | |
| "loss": 0.1151, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 1.1009399264405396e-05, | |
| "loss": 0.0985, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 1.0192071924805887e-05, | |
| "loss": 0.0941, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 9.374744585206375e-06, | |
| "loss": 0.1066, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 8.557417245606866e-06, | |
| "loss": 0.0922, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9910248986682108, | |
| "eval_loss": 0.034099191427230835, | |
| "eval_runtime": 239.3405, | |
| "eval_samples_per_second": 14.431, | |
| "eval_steps_per_second": 1.805, | |
| "step": 7341 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 7.740089906007356e-06, | |
| "loss": 0.091, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 6.922762566407847e-06, | |
| "loss": 0.0666, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 6.105435226808338e-06, | |
| "loss": 0.0667, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 5.288107887208827e-06, | |
| "loss": 0.0727, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "learning_rate": 4.470780547609318e-06, | |
| "loss": 0.0592, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9939200926462073, | |
| "eval_loss": 0.021338362246751785, | |
| "eval_runtime": 246.6506, | |
| "eval_samples_per_second": 14.004, | |
| "eval_steps_per_second": 1.751, | |
| "step": 9788 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 3.6534532080098086e-06, | |
| "loss": 0.0646, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "learning_rate": 2.8361258684102986e-06, | |
| "loss": 0.0601, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 2.018798528810789e-06, | |
| "loss": 0.0556, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 1.201471189211279e-06, | |
| "loss": 0.0506, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 3.841438496117696e-07, | |
| "loss": 0.053, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9942096120440069, | |
| "eval_loss": 0.02587772160768509, | |
| "eval_runtime": 241.5304, | |
| "eval_samples_per_second": 14.3, | |
| "eval_steps_per_second": 1.789, | |
| "step": 12235 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 12235, | |
| "total_flos": 2.6807470622410383e+19, | |
| "train_loss": 0.11058193139753197, | |
| "train_runtime": 9068.2818, | |
| "train_samples_per_second": 10.791, | |
| "train_steps_per_second": 1.349 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 12235, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 2.6807470622410383e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |