B4_1000_1e-5_hp-myself-2 / trainer_state.json
lnxdx's picture
End of training
2841ffe
raw
history blame
4.29 kB
{
"best_metric": 0.3130841121495327,
"best_model_checkpoint": "/content/drive/MyDrive/wav2vec2-large-xlsr-persian-asr-shemo_sm95/checkpoint-1000",
"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": 0.8083,
"step": 100
},
{
"epoch": 0.62,
"eval_loss": 0.6765980124473572,
"eval_runtime": 35.6119,
"eval_samples_per_second": 7.975,
"eval_steps_per_second": 1.011,
"eval_wer": 0.32710280373831774,
"step": 100
},
{
"epoch": 1.25,
"learning_rate": 3.980000000000001e-06,
"loss": 0.8414,
"step": 200
},
{
"epoch": 1.25,
"eval_loss": 0.677401602268219,
"eval_runtime": 34.1374,
"eval_samples_per_second": 8.319,
"eval_steps_per_second": 1.055,
"eval_wer": 0.32593457943925236,
"step": 200
},
{
"epoch": 1.88,
"learning_rate": 5.98e-06,
"loss": 0.8465,
"step": 300
},
{
"epoch": 1.88,
"eval_loss": 0.6685968041419983,
"eval_runtime": 34.2793,
"eval_samples_per_second": 8.285,
"eval_steps_per_second": 1.05,
"eval_wer": 0.3262266355140187,
"step": 300
},
{
"epoch": 2.5,
"learning_rate": 7.980000000000002e-06,
"loss": 0.7819,
"step": 400
},
{
"epoch": 2.5,
"eval_loss": 0.6748902797698975,
"eval_runtime": 33.8091,
"eval_samples_per_second": 8.4,
"eval_steps_per_second": 1.065,
"eval_wer": 0.32067757009345793,
"step": 400
},
{
"epoch": 3.12,
"learning_rate": 9.980000000000001e-06,
"loss": 0.7905,
"step": 500
},
{
"epoch": 3.12,
"eval_loss": 0.6847506165504456,
"eval_runtime": 34.8761,
"eval_samples_per_second": 8.143,
"eval_steps_per_second": 1.032,
"eval_wer": 0.3177570093457944,
"step": 500
},
{
"epoch": 3.75,
"learning_rate": 8.020000000000001e-06,
"loss": 0.8078,
"step": 600
},
{
"epoch": 3.75,
"eval_loss": 0.6571177840232849,
"eval_runtime": 34.5072,
"eval_samples_per_second": 8.23,
"eval_steps_per_second": 1.043,
"eval_wer": 0.3244742990654206,
"step": 600
},
{
"epoch": 4.38,
"learning_rate": 6.02e-06,
"loss": 0.7771,
"step": 700
},
{
"epoch": 4.38,
"eval_loss": 0.6683077812194824,
"eval_runtime": 34.8133,
"eval_samples_per_second": 8.158,
"eval_steps_per_second": 1.034,
"eval_wer": 0.31454439252336447,
"step": 700
},
{
"epoch": 5.0,
"learning_rate": 4.0200000000000005e-06,
"loss": 0.7786,
"step": 800
},
{
"epoch": 5.0,
"eval_loss": 0.668791651725769,
"eval_runtime": 37.4742,
"eval_samples_per_second": 7.579,
"eval_steps_per_second": 0.961,
"eval_wer": 0.31366822429906543,
"step": 800
},
{
"epoch": 5.62,
"learning_rate": 2.02e-06,
"loss": 0.7656,
"step": 900
},
{
"epoch": 5.62,
"eval_loss": 0.6702979803085327,
"eval_runtime": 35.4948,
"eval_samples_per_second": 8.001,
"eval_steps_per_second": 1.014,
"eval_wer": 0.3133761682242991,
"step": 900
},
{
"epoch": 6.25,
"learning_rate": 2e-08,
"loss": 0.7516,
"step": 1000
},
{
"epoch": 6.25,
"eval_loss": 0.6705625057220459,
"eval_runtime": 35.3775,
"eval_samples_per_second": 8.028,
"eval_steps_per_second": 1.018,
"eval_wer": 0.3130841121495327,
"step": 1000
},
{
"epoch": 6.25,
"step": 1000,
"total_flos": 4.798362622332561e+18,
"train_loss": 0.7949239425659179,
"train_runtime": 4775.7653,
"train_samples_per_second": 3.35,
"train_steps_per_second": 0.209
}
],
"logging_steps": 100,
"max_steps": 1000,
"num_train_epochs": 7,
"save_steps": 100,
"total_flos": 4.798362622332561e+18,
"trial_name": null,
"trial_params": null
}