Model-J__MAE__model_idx_0653 / trainer_state.json
Eliahu's picture
Upload folder using huggingface_hub
1cd1f4a verified
Invalid JSON: Unexpected token 'I', ..."ad_norm": Infinity, "... is not valid JSON
{
"best_metric": 0.8821333333333333,
"best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0653/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": Infinity,
"learning_rate": 9e-05,
"loss": 1.757,
"step": 333
},
{
"epoch": 1.0,
"eval_accuracy": 0.8216,
"eval_loss": 0.628304123878479,
"eval_runtime": 4.058,
"eval_samples_per_second": 924.094,
"eval_steps_per_second": 14.539,
"step": 333
},
{
"epoch": 2.0,
"grad_norm": Infinity,
"learning_rate": 9e-05,
"loss": 0.4632,
"step": 666
},
{
"epoch": 2.0,
"eval_accuracy": 0.8496,
"eval_loss": 0.4910394847393036,
"eval_runtime": 4.0013,
"eval_samples_per_second": 937.199,
"eval_steps_per_second": 14.745,
"step": 666
},
{
"epoch": 3.0,
"grad_norm": 4.267885684967041,
"learning_rate": 9e-05,
"loss": 0.2616,
"step": 999
},
{
"epoch": 3.0,
"eval_accuracy": 0.8573333333333333,
"eval_loss": 0.4930065870285034,
"eval_runtime": 4.0273,
"eval_samples_per_second": 931.135,
"eval_steps_per_second": 14.65,
"step": 999
},
{
"epoch": 4.0,
"grad_norm": 30.387842178344727,
"learning_rate": 9e-05,
"loss": 0.1841,
"step": 1332
},
{
"epoch": 4.0,
"eval_accuracy": 0.8821333333333333,
"eval_loss": 0.39944830536842346,
"eval_runtime": 4.0413,
"eval_samples_per_second": 927.909,
"eval_steps_per_second": 14.599,
"step": 1332
}
],
"logging_steps": 500,
"max_steps": 1998,
"num_input_tokens_seen": 0,
"num_train_epochs": 6,
"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
}