trainer_state.json

{
  "best_metric": 0.8613333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0708/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "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": 0.0001,
      "loss": 1.7116,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7946666666666666,
      "eval_loss": 0.7471078634262085,
      "eval_runtime": 7.4605,
      "eval_samples_per_second": 502.649,
      "eval_steps_per_second": 7.908,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.5816150903701782,
      "learning_rate": 0.0001,
      "loss": 0.5873,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8368,
      "eval_loss": 0.59164959192276,
      "eval_runtime": 8.0359,
      "eval_samples_per_second": 466.657,
      "eval_steps_per_second": 7.342,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.280129432678223,
      "learning_rate": 0.0001,
      "loss": 0.3678,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8602666666666666,
      "eval_loss": 0.48524022102355957,
      "eval_runtime": 7.4183,
      "eval_samples_per_second": 505.505,
      "eval_steps_per_second": 7.953,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0001,
      "loss": 0.2642,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8557333333333333,
      "eval_loss": 0.5309296250343323,
      "eval_runtime": 7.7882,
      "eval_samples_per_second": 481.497,
      "eval_steps_per_second": 7.576,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.2055439949035645,
      "learning_rate": 0.0001,
      "loss": 0.188,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8613333333333333,
      "eval_loss": 0.5082705616950989,
      "eval_runtime": 7.638,
      "eval_samples_per_second": 490.966,
      "eval_steps_per_second": 7.725,
      "step": 1665
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}