C1_1000_1e-5_mehrdad / trainer_state.json
lnxdx's picture
End of training
a9ceb3d
{
"best_metric": 0.3142523364485981,
"best_model_checkpoint": "/content/drive/MyDrive/wav2vec2-large-xlsr-persian-asr-shemo_sm95/checkpoint-900",
"epoch": 6.25,
"eval_steps": 100,
"global_step": 1000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.62,
"learning_rate": 2.0000000000000003e-06,
"loss": 0.7062,
"step": 100
},
{
"epoch": 0.62,
"eval_loss": 0.6594027876853943,
"eval_runtime": 34.2834,
"eval_samples_per_second": 8.284,
"eval_steps_per_second": 1.05,
"eval_wer": 0.31600467289719625,
"step": 100
},
{
"epoch": 1.25,
"learning_rate": 3.96e-06,
"loss": 0.7111,
"step": 200
},
{
"epoch": 1.25,
"eval_loss": 0.6545001864433289,
"eval_runtime": 34.16,
"eval_samples_per_second": 8.314,
"eval_steps_per_second": 1.054,
"eval_wer": 0.3177570093457944,
"step": 200
},
{
"epoch": 1.88,
"learning_rate": 5.9600000000000005e-06,
"loss": 0.7094,
"step": 300
},
{
"epoch": 1.88,
"eval_loss": 0.6424959301948547,
"eval_runtime": 33.6744,
"eval_samples_per_second": 8.434,
"eval_steps_per_second": 1.069,
"eval_wer": 0.32184579439252337,
"step": 300
},
{
"epoch": 2.5,
"learning_rate": 7.960000000000002e-06,
"loss": 0.649,
"step": 400
},
{
"epoch": 2.5,
"eval_loss": 0.6548773646354675,
"eval_runtime": 36.0564,
"eval_samples_per_second": 7.877,
"eval_steps_per_second": 0.998,
"eval_wer": 0.31863317757009346,
"step": 400
},
{
"epoch": 3.12,
"learning_rate": 9.960000000000001e-06,
"loss": 0.6436,
"step": 500
},
{
"epoch": 3.12,
"eval_loss": 0.6656894683837891,
"eval_runtime": 33.8076,
"eval_samples_per_second": 8.4,
"eval_steps_per_second": 1.065,
"eval_wer": 0.31542056074766356,
"step": 500
},
{
"epoch": 3.75,
"learning_rate": 8.040000000000001e-06,
"loss": 0.6541,
"step": 600
},
{
"epoch": 3.75,
"eval_loss": 0.6529704332351685,
"eval_runtime": 34.802,
"eval_samples_per_second": 8.16,
"eval_steps_per_second": 1.034,
"eval_wer": 0.3171728971962617,
"step": 600
},
{
"epoch": 4.38,
"learning_rate": 6.040000000000001e-06,
"loss": 0.6317,
"step": 700
},
{
"epoch": 4.38,
"eval_loss": 0.6600654125213623,
"eval_runtime": 34.964,
"eval_samples_per_second": 8.123,
"eval_steps_per_second": 1.03,
"eval_wer": 0.3151285046728972,
"step": 700
},
{
"epoch": 5.0,
"learning_rate": 4.060000000000001e-06,
"loss": 0.6405,
"step": 800
},
{
"epoch": 5.0,
"eval_loss": 0.658935010433197,
"eval_runtime": 34.6058,
"eval_samples_per_second": 8.207,
"eval_steps_per_second": 1.04,
"eval_wer": 0.3180490654205608,
"step": 800
},
{
"epoch": 5.62,
"learning_rate": 2.06e-06,
"loss": 0.6284,
"step": 900
},
{
"epoch": 5.62,
"eval_loss": 0.6640745401382446,
"eval_runtime": 35.8838,
"eval_samples_per_second": 7.914,
"eval_steps_per_second": 1.003,
"eval_wer": 0.3142523364485981,
"step": 900
},
{
"epoch": 6.25,
"learning_rate": 6.000000000000001e-08,
"loss": 0.6287,
"step": 1000
},
{
"epoch": 6.25,
"eval_loss": 0.662032425403595,
"eval_runtime": 34.5828,
"eval_samples_per_second": 8.212,
"eval_steps_per_second": 1.041,
"eval_wer": 0.3151285046728972,
"step": 1000
},
{
"epoch": 6.25,
"step": 1000,
"total_flos": 4.798362622332561e+18,
"train_loss": 0.6602616081237793,
"train_runtime": 4840.2702,
"train_samples_per_second": 3.306,
"train_steps_per_second": 0.207
}
],
"logging_steps": 100,
"max_steps": 1000,
"num_train_epochs": 7,
"save_steps": 100,
"total_flos": 4.798362622332561e+18,
"trial_name": null,
"trial_params": null
}