trainer_state.json

{
  "best_metric": 0.8474666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0579/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": 9e-05,
      "loss": 1.651,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.776,
      "eval_loss": 0.7881268262863159,
      "eval_runtime": 5.6445,
      "eval_samples_per_second": 664.368,
      "eval_steps_per_second": 10.453,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.6330879330635071,
      "learning_rate": 9e-05,
      "loss": 0.5176,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8128,
      "eval_loss": 0.6270188093185425,
      "eval_runtime": 5.4046,
      "eval_samples_per_second": 693.854,
      "eval_steps_per_second": 10.917,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.733447790145874,
      "learning_rate": 9e-05,
      "loss": 0.2742,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8184,
      "eval_loss": 0.6397126317024231,
      "eval_runtime": 5.711,
      "eval_samples_per_second": 656.629,
      "eval_steps_per_second": 10.331,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.1703767031431198,
      "learning_rate": 9e-05,
      "loss": 0.1582,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8304,
      "eval_loss": 0.6008048057556152,
      "eval_runtime": 5.4686,
      "eval_samples_per_second": 685.728,
      "eval_steps_per_second": 10.789,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.38539204001426697,
      "learning_rate": 9e-05,
      "loss": 0.1089,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8474666666666667,
      "eval_loss": 0.5811952948570251,
      "eval_runtime": 5.4182,
      "eval_samples_per_second": 692.107,
      "eval_steps_per_second": 10.889,
      "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
}