MESA / checkpoint-50 /trainer_state.json
Gitesh2003's picture
Upload 14 files
b4386c9 verified
{
"best_metric": 1.3204787969589233,
"best_model_checkpoint": "./trocr_finetuned\\checkpoint-50",
"epoch": 2.0,
"eval_steps": 500,
"global_step": 50,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.4,
"grad_norm": 107.13959503173828,
"learning_rate": 4e-05,
"loss": 2.2609,
"step": 10
},
{
"epoch": 0.8,
"grad_norm": 59.0462532043457,
"learning_rate": 3e-05,
"loss": 2.1178,
"step": 20
},
{
"epoch": 1.0,
"eval_loss": 1.566049337387085,
"eval_runtime": 62.707,
"eval_samples_per_second": 0.319,
"eval_steps_per_second": 0.08,
"step": 25
},
{
"epoch": 1.2,
"grad_norm": 25.392553329467773,
"learning_rate": 2e-05,
"loss": 1.1832,
"step": 30
},
{
"epoch": 1.6,
"grad_norm": 103.50560760498047,
"learning_rate": 1e-05,
"loss": 0.6477,
"step": 40
},
{
"epoch": 2.0,
"grad_norm": 27.466867446899414,
"learning_rate": 0.0,
"loss": 0.4053,
"step": 50
},
{
"epoch": 2.0,
"eval_loss": 1.3204787969589233,
"eval_runtime": 49.7756,
"eval_samples_per_second": 0.402,
"eval_steps_per_second": 0.1,
"step": 50
}
],
"logging_steps": 10,
"max_steps": 50,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 1.496570381991936e+17,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}