| { | |
| "best_metric": 1.0641573667526245, | |
| "best_model_checkpoint": "fairface_age_image_detection/checkpoint-9376", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 9376, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.21331058020477817, | |
| "grad_norm": 1.7275338172912598, | |
| "learning_rate": 2.855243405532919e-06, | |
| "loss": 1.8728, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.42662116040955633, | |
| "grad_norm": 2.228030204772949, | |
| "learning_rate": 2.6944027450139397e-06, | |
| "loss": 1.4714, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6399317406143344, | |
| "grad_norm": 2.2512168884277344, | |
| "learning_rate": 2.5335620844949606e-06, | |
| "loss": 1.3358, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.8532423208191127, | |
| "grad_norm": 3.124258279800415, | |
| "learning_rate": 2.372721423975981e-06, | |
| "loss": 1.264, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5423, | |
| "eval_loss": 1.2151408195495605, | |
| "eval_model_preparation_time": 0.0043, | |
| "eval_runtime": 907.0206, | |
| "eval_samples_per_second": 11.025, | |
| "eval_steps_per_second": 1.378, | |
| "step": 2344 | |
| }, | |
| { | |
| "epoch": 1.0665529010238908, | |
| "grad_norm": 5.8204874992370605, | |
| "learning_rate": 2.211880763457002e-06, | |
| "loss": 1.2174, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.2798634812286689, | |
| "grad_norm": 5.175824165344238, | |
| "learning_rate": 2.051040102938023e-06, | |
| "loss": 1.171, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.493174061433447, | |
| "grad_norm": 3.312744617462158, | |
| "learning_rate": 1.8901994424190437e-06, | |
| "loss": 1.1425, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.7064846416382253, | |
| "grad_norm": 4.630620002746582, | |
| "learning_rate": 1.7293587819000642e-06, | |
| "loss": 1.1267, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.9197952218430034, | |
| "grad_norm": 3.5600452423095703, | |
| "learning_rate": 1.5685181213810853e-06, | |
| "loss": 1.1051, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5688, | |
| "eval_loss": 1.1124917268753052, | |
| "eval_model_preparation_time": 0.0043, | |
| "eval_runtime": 915.0894, | |
| "eval_samples_per_second": 10.928, | |
| "eval_steps_per_second": 1.366, | |
| "step": 4688 | |
| }, | |
| { | |
| "epoch": 2.1331058020477816, | |
| "grad_norm": 7.453662395477295, | |
| "learning_rate": 1.407677460862106e-06, | |
| "loss": 1.0871, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.34641638225256, | |
| "grad_norm": 8.157800674438477, | |
| "learning_rate": 1.2468368003431269e-06, | |
| "loss": 1.0696, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.5597269624573378, | |
| "grad_norm": 3.6688616275787354, | |
| "learning_rate": 1.0859961398241475e-06, | |
| "loss": 1.0571, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.773037542662116, | |
| "grad_norm": 3.1099190711975098, | |
| "learning_rate": 9.251554793051683e-07, | |
| "loss": 1.0476, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.986348122866894, | |
| "grad_norm": 5.784822463989258, | |
| "learning_rate": 7.643148187861893e-07, | |
| "loss": 1.0407, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5756, | |
| "eval_loss": 1.0761935710906982, | |
| "eval_model_preparation_time": 0.0043, | |
| "eval_runtime": 920.0161, | |
| "eval_samples_per_second": 10.869, | |
| "eval_steps_per_second": 1.359, | |
| "step": 7032 | |
| }, | |
| { | |
| "epoch": 3.1996587030716723, | |
| "grad_norm": 5.856372833251953, | |
| "learning_rate": 6.0347415826721e-07, | |
| "loss": 1.0212, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.4129692832764507, | |
| "grad_norm": 5.725022792816162, | |
| "learning_rate": 4.426334977482307e-07, | |
| "loss": 1.0253, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.6262798634812285, | |
| "grad_norm": 4.74147891998291, | |
| "learning_rate": 2.8179283722925154e-07, | |
| "loss": 1.0232, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.839590443686007, | |
| "grad_norm": 4.050413131713867, | |
| "learning_rate": 1.2095217671027237e-07, | |
| "loss": 1.0173, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5791, | |
| "eval_loss": 1.0641573667526245, | |
| "eval_model_preparation_time": 0.0043, | |
| "eval_runtime": 919.3696, | |
| "eval_samples_per_second": 10.877, | |
| "eval_steps_per_second": 1.36, | |
| "step": 9376 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 9376, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "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": 2.32490554103808e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |