Model-J__SupViT__model_idx_0950 / trainer_state.json
Eliahu's picture
Upload folder using huggingface_hub
31c8905 verified
{
"best_metric": 0.9578666666666666,
"best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0950/checkpoints/checkpoint-1332",
"epoch": 4.0,
"eval_steps": 500,
"global_step": 1332,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"grad_norm": 26.928781509399414,
"learning_rate": 9.045084971874738e-05,
"loss": 0.6047,
"step": 333
},
{
"epoch": 1.0,
"eval_accuracy": 0.9245333333333333,
"eval_loss": 0.271129846572876,
"eval_runtime": 7.3581,
"eval_samples_per_second": 509.645,
"eval_steps_per_second": 8.018,
"step": 333
},
{
"epoch": 2.0,
"grad_norm": 0.07842057943344116,
"learning_rate": 6.545084971874738e-05,
"loss": 0.1303,
"step": 666
},
{
"epoch": 2.0,
"eval_accuracy": 0.9416,
"eval_loss": 0.21038730442523956,
"eval_runtime": 7.0409,
"eval_samples_per_second": 532.603,
"eval_steps_per_second": 8.38,
"step": 666
},
{
"epoch": 3.0,
"grad_norm": 0.06460259109735489,
"learning_rate": 3.4549150281252636e-05,
"loss": 0.043,
"step": 999
},
{
"epoch": 3.0,
"eval_accuracy": 0.9498666666666666,
"eval_loss": 0.1930592805147171,
"eval_runtime": 7.0389,
"eval_samples_per_second": 532.756,
"eval_steps_per_second": 8.382,
"step": 999
},
{
"epoch": 4.0,
"grad_norm": 0.030140871182084084,
"learning_rate": 9.549150281252633e-06,
"loss": 0.0106,
"step": 1332
},
{
"epoch": 4.0,
"eval_accuracy": 0.9578666666666666,
"eval_loss": 0.16957108676433563,
"eval_runtime": 7.1957,
"eval_samples_per_second": 521.145,
"eval_steps_per_second": 8.199,
"step": 1332
}
],
"logging_steps": 500,
"max_steps": 1665,
"num_input_tokens_seen": 0,
"num_train_epochs": 5,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 6.58965282490368e+18,
"train_batch_size": 64,
"trial_name": null,
"trial_params": null
}