Model-J__SupViT__model_idx_0679 / trainer_state.json
Eliahu's picture
Upload folder using huggingface_hub
41ea96b verified
{
"best_metric": 0.9448,
"best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0679/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": 0.692074716091156,
"learning_rate": 6.331559480312315e-05,
"loss": 0.6766,
"step": 333
},
{
"epoch": 1.0,
"eval_accuracy": 0.9157333333333333,
"eval_loss": 0.29635557532310486,
"eval_runtime": 12.1694,
"eval_samples_per_second": 308.149,
"eval_steps_per_second": 4.848,
"step": 333
},
{
"epoch": 2.0,
"grad_norm": 0.181728795170784,
"learning_rate": 4.581559480312316e-05,
"loss": 0.1017,
"step": 666
},
{
"epoch": 2.0,
"eval_accuracy": 0.9336,
"eval_loss": 0.2474767565727234,
"eval_runtime": 12.0829,
"eval_samples_per_second": 310.355,
"eval_steps_per_second": 4.883,
"step": 666
},
{
"epoch": 3.0,
"grad_norm": 0.18909093737602234,
"learning_rate": 2.4184405196876842e-05,
"loss": 0.0242,
"step": 999
},
{
"epoch": 3.0,
"eval_accuracy": 0.9418666666666666,
"eval_loss": 0.22181198000907898,
"eval_runtime": 11.9853,
"eval_samples_per_second": 312.882,
"eval_steps_per_second": 4.923,
"step": 999
},
{
"epoch": 4.0,
"grad_norm": 0.03488137573003769,
"learning_rate": 6.684405196876842e-06,
"loss": 0.0072,
"step": 1332
},
{
"epoch": 4.0,
"eval_accuracy": 0.9448,
"eval_loss": 0.2141786813735962,
"eval_runtime": 11.8703,
"eval_samples_per_second": 315.914,
"eval_steps_per_second": 4.97,
"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
}