| { | |
| "best_metric": 0.7416653633117676, | |
| "best_model_checkpoint": "face_emotions_image_detection/checkpoint-2380", | |
| "epoch": 7.0, | |
| "eval_steps": 500, | |
| "global_step": 2380, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7815473727761688, | |
| "eval_loss": 0.7615271210670471, | |
| "eval_runtime": 35.3314, | |
| "eval_samples_per_second": 68.409, | |
| "eval_steps_per_second": 2.151, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "grad_norm": 2.714886426925659, | |
| "learning_rate": 2.4206008583690987e-07, | |
| "loss": 0.6933, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.78113363673976, | |
| "eval_loss": 0.7570552825927734, | |
| "eval_runtime": 35.8902, | |
| "eval_samples_per_second": 67.344, | |
| "eval_steps_per_second": 2.118, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "grad_norm": 2.760270595550537, | |
| "learning_rate": 1.7768240343347639e-07, | |
| "loss": 0.6836, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7807199007033513, | |
| "eval_loss": 0.7524998188018799, | |
| "eval_runtime": 35.2486, | |
| "eval_samples_per_second": 68.57, | |
| "eval_steps_per_second": 2.156, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7852709971038477, | |
| "eval_loss": 0.7466434240341187, | |
| "eval_runtime": 35.4191, | |
| "eval_samples_per_second": 68.24, | |
| "eval_steps_per_second": 2.146, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "grad_norm": 2.9822440147399902, | |
| "learning_rate": 1.1330472103004292e-07, | |
| "loss": 0.6746, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7860984691766653, | |
| "eval_loss": 0.7431991696357727, | |
| "eval_runtime": 35.5228, | |
| "eval_samples_per_second": 68.041, | |
| "eval_steps_per_second": 2.139, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "grad_norm": 2.6914942264556885, | |
| "learning_rate": 4.8927038626609435e-08, | |
| "loss": 0.6715, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7877534133223004, | |
| "eval_loss": 0.7418521642684937, | |
| "eval_runtime": 34.9484, | |
| "eval_samples_per_second": 69.159, | |
| "eval_steps_per_second": 2.175, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7869259412494828, | |
| "eval_loss": 0.7416653633117676, | |
| "eval_runtime": 34.4685, | |
| "eval_samples_per_second": 70.122, | |
| "eval_steps_per_second": 2.205, | |
| "step": 2380 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2380, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "total_flos": 1.1795866074490982e+19, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |