cdetr-r50-cd45rb-all-4ah / trainer_state.json
polejowska's picture
End of training
681580b
raw
history blame
7.13 kB
{
"best_metric": 1.7968701124191284,
"best_model_checkpoint": "cdetr-r50-cd45rb-all-4ah\\checkpoint-46060",
"epoch": 20.0,
"global_step": 46060,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"learning_rate": 9.003039513677813e-06,
"loss": 2.165,
"step": 2303
},
{
"epoch": 1.0,
"eval_loss": 2.0256807804107666,
"eval_runtime": 417.1917,
"eval_samples_per_second": 4.269,
"eval_steps_per_second": 0.535,
"step": 2303
},
{
"epoch": 2.0,
"learning_rate": 8.003907946157187e-06,
"loss": 1.9843,
"step": 4606
},
{
"epoch": 2.0,
"eval_loss": 1.9741652011871338,
"eval_runtime": 403.5768,
"eval_samples_per_second": 4.413,
"eval_steps_per_second": 0.553,
"step": 4606
},
{
"epoch": 3.0,
"learning_rate": 7.003907946157187e-06,
"loss": 1.9296,
"step": 6909
},
{
"epoch": 3.0,
"eval_loss": 1.9122804403305054,
"eval_runtime": 401.3575,
"eval_samples_per_second": 4.437,
"eval_steps_per_second": 0.556,
"step": 6909
},
{
"epoch": 4.0,
"learning_rate": 6.0043421623968744e-06,
"loss": 1.8949,
"step": 9212
},
{
"epoch": 4.0,
"eval_loss": 1.8927873373031616,
"eval_runtime": 408.1494,
"eval_samples_per_second": 4.364,
"eval_steps_per_second": 0.546,
"step": 9212
},
{
"epoch": 5.0,
"learning_rate": 5.005210594876249e-06,
"loss": 1.8751,
"step": 11515
},
{
"epoch": 5.0,
"eval_loss": 1.9116449356079102,
"eval_runtime": 406.0082,
"eval_samples_per_second": 4.387,
"eval_steps_per_second": 0.549,
"step": 11515
},
{
"epoch": 6.0,
"learning_rate": 4.005210594876249e-06,
"loss": 1.854,
"step": 13818
},
{
"epoch": 6.0,
"eval_loss": 1.8909248113632202,
"eval_runtime": 396.4567,
"eval_samples_per_second": 4.492,
"eval_steps_per_second": 0.562,
"step": 13818
},
{
"epoch": 7.0,
"learning_rate": 3.005644811115936e-06,
"loss": 1.8354,
"step": 16121
},
{
"epoch": 7.0,
"eval_loss": 1.8532779216766357,
"eval_runtime": 390.3322,
"eval_samples_per_second": 4.563,
"eval_steps_per_second": 0.571,
"step": 16121
},
{
"epoch": 8.0,
"learning_rate": 2.0060790273556234e-06,
"loss": 1.8257,
"step": 18424
},
{
"epoch": 8.0,
"eval_loss": 1.8406795263290405,
"eval_runtime": 388.3033,
"eval_samples_per_second": 4.587,
"eval_steps_per_second": 0.574,
"step": 18424
},
{
"epoch": 9.0,
"learning_rate": 1.0060790273556232e-06,
"loss": 1.8101,
"step": 20727
},
{
"epoch": 9.0,
"eval_loss": 1.8452762365341187,
"eval_runtime": 395.2037,
"eval_samples_per_second": 4.507,
"eval_steps_per_second": 0.564,
"step": 20727
},
{
"epoch": 10.0,
"learning_rate": 6.947459834997829e-09,
"loss": 1.8025,
"step": 23030
},
{
"epoch": 10.0,
"eval_loss": 1.8343534469604492,
"eval_runtime": 390.5312,
"eval_samples_per_second": 4.56,
"eval_steps_per_second": 0.571,
"step": 23030
},
{
"epoch": 11.0,
"learning_rate": 4.503473729917499e-06,
"loss": 1.8177,
"step": 25333
},
{
"epoch": 11.0,
"eval_loss": 1.8496007919311523,
"eval_runtime": 388.6824,
"eval_samples_per_second": 4.582,
"eval_steps_per_second": 0.574,
"step": 25333
},
{
"epoch": 12.0,
"learning_rate": 4.003690838037343e-06,
"loss": 1.8092,
"step": 27636
},
{
"epoch": 12.0,
"eval_loss": 1.8583861589431763,
"eval_runtime": 382.9251,
"eval_samples_per_second": 4.651,
"eval_steps_per_second": 0.582,
"step": 27636
},
{
"epoch": 13.0,
"learning_rate": 3.5039079461571867e-06,
"loss": 1.8016,
"step": 29939
},
{
"epoch": 13.0,
"eval_loss": 1.8277424573898315,
"eval_runtime": 375.0626,
"eval_samples_per_second": 4.749,
"eval_steps_per_second": 0.595,
"step": 29939
},
{
"epoch": 14.0,
"learning_rate": 3.0039079461571864e-06,
"loss": 1.7961,
"step": 32242
},
{
"epoch": 14.0,
"eval_loss": 1.8229725360870361,
"eval_runtime": 215.0084,
"eval_samples_per_second": 8.283,
"eval_steps_per_second": 1.037,
"step": 32242
},
{
"epoch": 15.0,
"learning_rate": 2.50412505427703e-06,
"loss": 1.7853,
"step": 34545
},
{
"epoch": 15.0,
"eval_loss": 1.8130303621292114,
"eval_runtime": 207.7071,
"eval_samples_per_second": 8.575,
"eval_steps_per_second": 1.074,
"step": 34545
},
{
"epoch": 16.0,
"learning_rate": 2.0043421623968738e-06,
"loss": 1.7769,
"step": 36848
},
{
"epoch": 16.0,
"eval_loss": 1.8098373413085938,
"eval_runtime": 212.184,
"eval_samples_per_second": 8.394,
"eval_steps_per_second": 1.051,
"step": 36848
},
{
"epoch": 17.0,
"learning_rate": 1.5045592705167174e-06,
"loss": 1.7708,
"step": 39151
},
{
"epoch": 17.0,
"eval_loss": 1.8029736280441284,
"eval_runtime": 208.316,
"eval_samples_per_second": 8.55,
"eval_steps_per_second": 1.07,
"step": 39151
},
{
"epoch": 18.0,
"learning_rate": 1.0049934867564047e-06,
"loss": 1.7626,
"step": 41454
},
{
"epoch": 18.0,
"eval_loss": 1.7993708848953247,
"eval_runtime": 207.7204,
"eval_samples_per_second": 8.574,
"eval_steps_per_second": 1.074,
"step": 41454
},
{
"epoch": 19.0,
"learning_rate": 5.049934867564047e-07,
"loss": 1.7596,
"step": 43757
},
{
"epoch": 19.0,
"eval_loss": 1.8018128871917725,
"eval_runtime": 211.9911,
"eval_samples_per_second": 8.401,
"eval_steps_per_second": 1.052,
"step": 43757
},
{
"epoch": 20.0,
"learning_rate": 5.210594876248372e-09,
"loss": 1.7547,
"step": 46060
},
{
"epoch": 20.0,
"eval_loss": 1.7968701124191284,
"eval_runtime": 439.0653,
"eval_samples_per_second": 4.056,
"eval_steps_per_second": 0.508,
"step": 46060
},
{
"epoch": 20.0,
"step": 46060,
"total_flos": 1.838847807523584e+20,
"train_loss": 0.8917218147440159,
"train_runtime": 42292.62,
"train_samples_per_second": 8.711,
"train_steps_per_second": 1.089
}
],
"max_steps": 46060,
"num_train_epochs": 20,
"total_flos": 1.838847807523584e+20,
"trial_name": null,
"trial_params": null
}