trainer_state.json

{
  "best_metric": 0.8936,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0866/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "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": 2.667667667667668e-05,
      "loss": 2.4263,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7389333333333333,
      "eval_loss": 1.1925911903381348,
      "eval_runtime": 13.6696,
      "eval_samples_per_second": 274.331,
      "eval_steps_per_second": 4.316,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 53.424522399902344,
      "learning_rate": 2.3343343343343343e-05,
      "loss": 0.8015,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8213333333333334,
      "eval_loss": 0.6646834015846252,
      "eval_runtime": 13.0554,
      "eval_samples_per_second": 287.237,
      "eval_steps_per_second": 4.519,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 2.002002002002002e-05,
      "loss": 0.4674,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8522666666666666,
      "eval_loss": 0.5173524618148804,
      "eval_runtime": 13.6595,
      "eval_samples_per_second": 274.534,
      "eval_steps_per_second": 4.319,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 34.578182220458984,
      "learning_rate": 1.668668668668669e-05,
      "loss": 0.3163,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8653333333333333,
      "eval_loss": 0.44937294721603394,
      "eval_runtime": 13.4831,
      "eval_samples_per_second": 278.126,
      "eval_steps_per_second": 4.376,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.5477373003959656,
      "learning_rate": 1.3353353353353355e-05,
      "loss": 0.2293,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8792,
      "eval_loss": 0.40815800428390503,
      "eval_runtime": 13.4538,
      "eval_samples_per_second": 278.731,
      "eval_steps_per_second": 4.385,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 15.978788375854492,
      "learning_rate": 1.002002002002002e-05,
      "loss": 0.1657,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8794666666666666,
      "eval_loss": 0.38971713185310364,
      "eval_runtime": 12.9159,
      "eval_samples_per_second": 290.34,
      "eval_steps_per_second": 4.568,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 1.3591634035110474,
      "learning_rate": 6.686686686686687e-06,
      "loss": 0.1208,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8864,
      "eval_loss": 0.3852943181991577,
      "eval_runtime": 13.534,
      "eval_samples_per_second": 277.081,
      "eval_steps_per_second": 4.359,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.4465229511260986,
      "learning_rate": 3.353353353353354e-06,
      "loss": 0.0892,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8936,
      "eval_loss": 0.36792126297950745,
      "eval_runtime": 13.4203,
      "eval_samples_per_second": 279.427,
      "eval_steps_per_second": 4.396,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}