trainer_state.json

{
  "best_metric": 0.8856,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0989/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "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": 0.0002887169311775794,
      "loss": 1.7807,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7269333333333333,
      "eval_loss": 0.9369990825653076,
      "eval_runtime": 6.8156,
      "eval_samples_per_second": 550.209,
      "eval_steps_per_second": 8.657,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 15.331551551818848,
      "learning_rate": 0.00025631588426322575,
      "loss": 0.6764,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8026666666666666,
      "eval_loss": 0.6695280075073242,
      "eval_runtime": 5.8325,
      "eval_samples_per_second": 642.947,
      "eval_steps_per_second": 10.116,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 6.320586681365967,
      "learning_rate": 0.0002077292077256867,
      "loss": 0.3988,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8146666666666667,
      "eval_loss": 0.6193615198135376,
      "eval_runtime": 5.8503,
      "eval_samples_per_second": 640.997,
      "eval_steps_per_second": 10.085,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 0.00015053067332772077,
      "loss": 0.227,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8264,
      "eval_loss": 0.586152195930481,
      "eval_runtime": 5.7954,
      "eval_samples_per_second": 647.063,
      "eval_steps_per_second": 10.18,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 20.570817947387695,
      "learning_rate": 9.308812260256249e-05,
      "loss": 0.1092,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8488,
      "eval_loss": 0.5494715571403503,
      "eval_runtime": 6.3153,
      "eval_samples_per_second": 593.796,
      "eval_steps_per_second": 9.342,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.09987344592809677,
      "learning_rate": 4.4309889303710674e-05,
      "loss": 0.0375,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8738666666666667,
      "eval_loss": 0.4741523563861847,
      "eval_runtime": 5.9877,
      "eval_samples_per_second": 626.284,
      "eval_steps_per_second": 9.854,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.0101927500218153,
      "learning_rate": 1.1622017275129708e-05,
      "loss": 0.0121,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8856,
      "eval_loss": 0.4502934217453003,
      "eval_runtime": 5.7629,
      "eval_samples_per_second": 650.713,
      "eval_steps_per_second": 10.238,
      "step": 2331
    }
  ],
  "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": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}