|
{ |
|
"best_metric": 18.367063400175983, |
|
"best_model_checkpoint": "./whisper-small-nl/checkpoint-600", |
|
"epoch": 1.194, |
|
"global_step": 1000, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.1, |
|
"learning_rate": 9.473684210526315e-06, |
|
"loss": 0.8378, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"eval_loss": 0.4932844042778015, |
|
"eval_runtime": 153.0661, |
|
"eval_samples_per_second": 6.533, |
|
"eval_steps_per_second": 0.105, |
|
"eval_wer": 23.882739776779513, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"learning_rate": 8.421052631578948e-06, |
|
"loss": 0.5547, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"eval_loss": 0.44761189818382263, |
|
"eval_runtime": 154.0782, |
|
"eval_samples_per_second": 6.49, |
|
"eval_steps_per_second": 0.104, |
|
"eval_wer": 21.057750196823044, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"learning_rate": 7.368421052631579e-06, |
|
"loss": 0.3905, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"eval_loss": 0.4335392117500305, |
|
"eval_runtime": 152.7461, |
|
"eval_samples_per_second": 6.547, |
|
"eval_steps_per_second": 0.105, |
|
"eval_wer": 21.168897327837723, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"learning_rate": 6.31578947368421e-06, |
|
"loss": 0.3766, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"eval_loss": 0.4266929030418396, |
|
"eval_runtime": 150.336, |
|
"eval_samples_per_second": 6.652, |
|
"eval_steps_per_second": 0.106, |
|
"eval_wer": 20.052794887231972, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.5, |
|
"learning_rate": 5.263157894736842e-06, |
|
"loss": 0.4164, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.5, |
|
"eval_loss": 0.4138640761375427, |
|
"eval_runtime": 157.7477, |
|
"eval_samples_per_second": 6.339, |
|
"eval_steps_per_second": 0.101, |
|
"eval_wer": 21.43287176399759, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.6, |
|
"learning_rate": 4.210526315789474e-06, |
|
"loss": 0.2939, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.6, |
|
"eval_loss": 0.38643914461135864, |
|
"eval_runtime": 151.6414, |
|
"eval_samples_per_second": 6.595, |
|
"eval_steps_per_second": 0.106, |
|
"eval_wer": 18.367063400175983, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"learning_rate": 3.157894736842105e-06, |
|
"loss": 0.2632, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"eval_loss": 0.3864086866378784, |
|
"eval_runtime": 152.5897, |
|
"eval_samples_per_second": 6.554, |
|
"eval_steps_per_second": 0.105, |
|
"eval_wer": 18.431899226601214, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"learning_rate": 2.105263157894737e-06, |
|
"loss": 0.6066, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"eval_loss": 0.3804214298725128, |
|
"eval_runtime": 156.1258, |
|
"eval_samples_per_second": 6.405, |
|
"eval_steps_per_second": 0.102, |
|
"eval_wer": 19.27476497012921, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 1.09, |
|
"learning_rate": 1.0526315789473685e-06, |
|
"loss": 0.2075, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 1.09, |
|
"eval_loss": 0.3793680667877197, |
|
"eval_runtime": 153.2522, |
|
"eval_samples_per_second": 6.525, |
|
"eval_steps_per_second": 0.104, |
|
"eval_wer": 18.890381142036773, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 1.19, |
|
"learning_rate": 0.0, |
|
"loss": 0.2102, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 1.19, |
|
"eval_loss": 0.37768101692199707, |
|
"eval_runtime": 153.5387, |
|
"eval_samples_per_second": 6.513, |
|
"eval_steps_per_second": 0.104, |
|
"eval_wer": 19.88144306025101, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 1.19, |
|
"step": 1000, |
|
"total_flos": 1.84579222044672e+19, |
|
"train_loss": 0.4157425765991211, |
|
"train_runtime": 7330.7001, |
|
"train_samples_per_second": 69.843, |
|
"train_steps_per_second": 0.136 |
|
} |
|
], |
|
"max_steps": 1000, |
|
"num_train_epochs": 9223372036854775807, |
|
"total_flos": 1.84579222044672e+19, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|