| { | |
| "best_metric": 0.06790149211883545, | |
| "best_model_checkpoint": "gender_face_image_detection/checkpoint-7956", | |
| "epoch": 9.0, | |
| "eval_steps": 500, | |
| "global_step": 7956, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.744027303754267e-06, | |
| "loss": 0.2517, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9735576923076923, | |
| "eval_loss": 0.09853074699640274, | |
| "eval_runtime": 70.669, | |
| "eval_samples_per_second": 100.072, | |
| "eval_steps_per_second": 12.509, | |
| "step": 884 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 4.459613196814562e-06, | |
| "loss": 0.0926, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 4.175199089874858e-06, | |
| "loss": 0.0831, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9789309954751131, | |
| "eval_loss": 0.08093170076608658, | |
| "eval_runtime": 69.4399, | |
| "eval_samples_per_second": 101.843, | |
| "eval_steps_per_second": 12.73, | |
| "step": 1768 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 3.890784982935154e-06, | |
| "loss": 0.0699, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 3.60637087599545e-06, | |
| "loss": 0.0628, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9809106334841629, | |
| "eval_loss": 0.07376975566148758, | |
| "eval_runtime": 68.4743, | |
| "eval_samples_per_second": 103.28, | |
| "eval_steps_per_second": 12.91, | |
| "step": 2652 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 3.3219567690557453e-06, | |
| "loss": 0.057, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 3.0375426621160415e-06, | |
| "loss": 0.0513, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9789309954751131, | |
| "eval_loss": 0.07511751353740692, | |
| "eval_runtime": 67.7191, | |
| "eval_samples_per_second": 104.431, | |
| "eval_steps_per_second": 13.054, | |
| "step": 3536 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 2.753128555176337e-06, | |
| "loss": 0.0444, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9811934389140271, | |
| "eval_loss": 0.06916382908821106, | |
| "eval_runtime": 69.275, | |
| "eval_samples_per_second": 102.086, | |
| "eval_steps_per_second": 12.761, | |
| "step": 4420 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "learning_rate": 2.4687144482366327e-06, | |
| "loss": 0.043, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 2.1843003412969285e-06, | |
| "loss": 0.0396, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9816176470588235, | |
| "eval_loss": 0.06955502182245255, | |
| "eval_runtime": 68.4827, | |
| "eval_samples_per_second": 103.267, | |
| "eval_steps_per_second": 12.908, | |
| "step": 5304 | |
| }, | |
| { | |
| "epoch": 6.22, | |
| "learning_rate": 1.8998862343572241e-06, | |
| "loss": 0.0394, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 6.79, | |
| "learning_rate": 1.61547212741752e-06, | |
| "loss": 0.0369, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9813348416289592, | |
| "eval_loss": 0.07007041573524475, | |
| "eval_runtime": 68.4061, | |
| "eval_samples_per_second": 103.383, | |
| "eval_steps_per_second": 12.923, | |
| "step": 6188 | |
| }, | |
| { | |
| "epoch": 7.35, | |
| "learning_rate": 1.331058020477816e-06, | |
| "loss": 0.0326, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 7.92, | |
| "learning_rate": 1.0466439135381116e-06, | |
| "loss": 0.0324, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9817590497737556, | |
| "eval_loss": 0.06906911730766296, | |
| "eval_runtime": 70.5486, | |
| "eval_samples_per_second": 100.243, | |
| "eval_steps_per_second": 12.53, | |
| "step": 7072 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "learning_rate": 7.622298065984074e-07, | |
| "loss": 0.0318, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9816176470588235, | |
| "eval_loss": 0.06790149211883545, | |
| "eval_runtime": 68.1765, | |
| "eval_samples_per_second": 103.731, | |
| "eval_steps_per_second": 12.966, | |
| "step": 7956 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 8840, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 1.972605090833208e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |