roberta-finetuned-squad_v2 / trainer_state.json
jonathanagustin's picture
Training in progress, step 1100
6fcd442
raw
history blame
4.22 kB
{
"best_metric": 0.9063502550125122,
"best_model_checkpoint": "/content/drive/My Drive/Colab Notebooks/aai520-project/checkpoints/roberta-finetuned/checkpoint-1000",
"epoch": 1.9607843137254903,
"eval_steps": 100,
"global_step": 1000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.2,
"learning_rate": 1.811764705882353e-05,
"loss": 2.9129,
"step": 100
},
{
"epoch": 0.2,
"eval_loss": 1.4699586629867554,
"eval_runtime": 17.6815,
"eval_samples_per_second": 676.131,
"eval_steps_per_second": 10.576,
"step": 100
},
{
"epoch": 0.39,
"learning_rate": 1.6176470588235296e-05,
"loss": 1.4395,
"step": 200
},
{
"epoch": 0.39,
"eval_loss": 1.240740418434143,
"eval_runtime": 17.6813,
"eval_samples_per_second": 676.14,
"eval_steps_per_second": 10.576,
"step": 200
},
{
"epoch": 0.59,
"learning_rate": 1.4215686274509805e-05,
"loss": 1.2356,
"step": 300
},
{
"epoch": 0.59,
"eval_loss": 1.0324599742889404,
"eval_runtime": 17.7184,
"eval_samples_per_second": 674.723,
"eval_steps_per_second": 10.554,
"step": 300
},
{
"epoch": 0.78,
"learning_rate": 1.2254901960784315e-05,
"loss": 1.1284,
"step": 400
},
{
"epoch": 0.78,
"eval_loss": 0.9749970436096191,
"eval_runtime": 17.7173,
"eval_samples_per_second": 674.763,
"eval_steps_per_second": 10.555,
"step": 400
},
{
"epoch": 0.98,
"learning_rate": 1.0294117647058823e-05,
"loss": 1.0821,
"step": 500
},
{
"epoch": 0.98,
"eval_loss": 0.9345471858978271,
"eval_runtime": 17.7101,
"eval_samples_per_second": 675.038,
"eval_steps_per_second": 10.559,
"step": 500
},
{
"epoch": 1.18,
"learning_rate": 8.333333333333334e-06,
"loss": 0.9978,
"step": 600
},
{
"epoch": 1.18,
"eval_loss": 0.9892988801002502,
"eval_runtime": 17.6864,
"eval_samples_per_second": 675.943,
"eval_steps_per_second": 10.573,
"step": 600
},
{
"epoch": 1.37,
"learning_rate": 6.372549019607843e-06,
"loss": 0.9697,
"step": 700
},
{
"epoch": 1.37,
"eval_loss": 0.9299613237380981,
"eval_runtime": 17.7225,
"eval_samples_per_second": 674.565,
"eval_steps_per_second": 10.552,
"step": 700
},
{
"epoch": 1.57,
"learning_rate": 4.411764705882353e-06,
"loss": 0.9455,
"step": 800
},
{
"epoch": 1.57,
"eval_loss": 0.9351199865341187,
"eval_runtime": 17.7361,
"eval_samples_per_second": 674.051,
"eval_steps_per_second": 10.543,
"step": 800
},
{
"epoch": 1.76,
"learning_rate": 2.450980392156863e-06,
"loss": 0.9322,
"step": 900
},
{
"epoch": 1.76,
"eval_loss": 0.9450510144233704,
"eval_runtime": 17.7032,
"eval_samples_per_second": 675.301,
"eval_steps_per_second": 10.563,
"step": 900
},
{
"epoch": 1.96,
"learning_rate": 4.901960784313725e-07,
"loss": 0.9269,
"step": 1000
},
{
"epoch": 1.96,
"eval_loss": 0.9063502550125122,
"eval_runtime": 17.6937,
"eval_samples_per_second": 675.663,
"eval_steps_per_second": 10.569,
"step": 1000
},
{
"epoch": 1.96,
"step": 1000,
"total_flos": 6.688961805360538e+16,
"train_loss": 0.0,
"train_runtime": 2.2787,
"train_samples_per_second": 229166.609,
"train_steps_per_second": 222.933
},
{
"epoch": 1.96,
"eval_loss": 0.9063556790351868,
"eval_runtime": 17.1279,
"eval_samples_per_second": 697.982,
"eval_steps_per_second": 2.744,
"step": 1000
}
],
"logging_steps": 100,
"max_steps": 508,
"num_train_epochs": 4,
"save_steps": 100,
"total_flos": 6.688961805360538e+16,
"trial_name": null,
"trial_params": null
}