trainer_state.json

{
  "best_metric": 0.8682666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0049/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": 8.696739196777344,
      "learning_rate": 0.0003,
      "loss": 0.5894,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8488,
      "eval_loss": 0.48898807168006897,
      "eval_runtime": 12.4084,
      "eval_samples_per_second": 302.214,
      "eval_steps_per_second": 4.755,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0003,
      "loss": 0.2878,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8565333333333334,
      "eval_loss": 0.4774414598941803,
      "eval_runtime": 12.8546,
      "eval_samples_per_second": 291.725,
      "eval_steps_per_second": 4.59,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.02356421761214733,
      "learning_rate": 0.0003,
      "loss": 0.211,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8368,
      "eval_loss": 0.6153307557106018,
      "eval_runtime": 12.9408,
      "eval_samples_per_second": 289.781,
      "eval_steps_per_second": 4.559,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.020769786089658737,
      "learning_rate": 0.0003,
      "loss": 0.1596,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8616,
      "eval_loss": 0.5148261785507202,
      "eval_runtime": 12.1965,
      "eval_samples_per_second": 307.464,
      "eval_steps_per_second": 4.837,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.08317369222640991,
      "learning_rate": 0.0003,
      "loss": 0.1458,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8682666666666666,
      "eval_loss": 0.5135239362716675,
      "eval_runtime": 12.3958,
      "eval_samples_per_second": 302.523,
      "eval_steps_per_second": 4.76,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}