trainer_state.json

{
  "best_metric": 0.788,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0523/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": 0.0003,
      "loss": 1.9757,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7045333333333333,
      "eval_loss": 0.9786065220832825,
      "eval_runtime": 11.8891,
      "eval_samples_per_second": 315.415,
      "eval_steps_per_second": 4.963,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 5.6521735191345215,
      "learning_rate": 0.0003,
      "loss": 0.7349,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7384,
      "eval_loss": 0.8590325713157654,
      "eval_runtime": 11.8057,
      "eval_samples_per_second": 317.644,
      "eval_steps_per_second": 4.998,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.6961029171943665,
      "learning_rate": 0.0003,
      "loss": 0.4436,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7744,
      "eval_loss": 0.7673046588897705,
      "eval_runtime": 11.8001,
      "eval_samples_per_second": 317.793,
      "eval_steps_per_second": 5.0,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.3282887935638428,
      "learning_rate": 0.0003,
      "loss": 0.2892,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7677333333333334,
      "eval_loss": 0.8282411694526672,
      "eval_runtime": 11.8463,
      "eval_samples_per_second": 316.555,
      "eval_steps_per_second": 4.98,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 10.16533374786377,
      "learning_rate": 0.0003,
      "loss": 0.2201,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.788,
      "eval_loss": 0.7695518136024475,
      "eval_runtime": 11.9082,
      "eval_samples_per_second": 314.909,
      "eval_steps_per_second": 4.955,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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
}