trainer_state.json

{
  "best_metric": 0.8965333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0170/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.9555903673171997,
      "learning_rate": 5e-05,
      "loss": 1.6878,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8213333333333334,
      "eval_loss": 0.7030234336853027,
      "eval_runtime": 11.9596,
      "eval_samples_per_second": 313.556,
      "eval_steps_per_second": 4.933,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 12.445852279663086,
      "learning_rate": 5e-05,
      "loss": 0.4934,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8533333333333334,
      "eval_loss": 0.49574920535087585,
      "eval_runtime": 11.3627,
      "eval_samples_per_second": 330.027,
      "eval_steps_per_second": 5.192,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 7.468923091888428,
      "learning_rate": 5e-05,
      "loss": 0.2881,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8829333333333333,
      "eval_loss": 0.4118095636367798,
      "eval_runtime": 10.6414,
      "eval_samples_per_second": 352.397,
      "eval_steps_per_second": 5.544,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 11.58633804321289,
      "learning_rate": 5e-05,
      "loss": 0.1937,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8872,
      "eval_loss": 0.40273594856262207,
      "eval_runtime": 10.3489,
      "eval_samples_per_second": 362.358,
      "eval_steps_per_second": 5.701,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 10.278923988342285,
      "learning_rate": 5e-05,
      "loss": 0.1268,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8890666666666667,
      "eval_loss": 0.4058311879634857,
      "eval_runtime": 10.7446,
      "eval_samples_per_second": 349.012,
      "eval_steps_per_second": 5.491,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.40097880363464355,
      "learning_rate": 5e-05,
      "loss": 0.0965,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8965333333333333,
      "eval_loss": 0.3907693326473236,
      "eval_runtime": 10.9817,
      "eval_samples_per_second": 341.477,
      "eval_steps_per_second": 5.373,
      "step": 1998
    }
  ],
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}