trainer_state.json

{
  "best_metric": 0.8650666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0660/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": 5e-05,
      "loss": 2.0645,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7450666666666667,
      "eval_loss": 0.9816085696220398,
      "eval_runtime": 6.3551,
      "eval_samples_per_second": 590.073,
      "eval_steps_per_second": 9.284,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 5e-05,
      "loss": 0.6998,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8173333333333334,
      "eval_loss": 0.6294081807136536,
      "eval_runtime": 7.2816,
      "eval_samples_per_second": 514.998,
      "eval_steps_per_second": 8.103,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.6507842540740967,
      "learning_rate": 5e-05,
      "loss": 0.4184,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8466666666666667,
      "eval_loss": 0.5329132080078125,
      "eval_runtime": 7.6307,
      "eval_samples_per_second": 491.434,
      "eval_steps_per_second": 7.732,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.8414950370788574,
      "learning_rate": 5e-05,
      "loss": 0.2929,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8450666666666666,
      "eval_loss": 0.509128987789154,
      "eval_runtime": 6.1464,
      "eval_samples_per_second": 610.115,
      "eval_steps_per_second": 9.599,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.3606182932853699,
      "learning_rate": 5e-05,
      "loss": 0.2137,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8650666666666667,
      "eval_loss": 0.4505220651626587,
      "eval_runtime": 6.5148,
      "eval_samples_per_second": 575.613,
      "eval_steps_per_second": 9.056,
      "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
}