trainer_state.json

{
  "best_metric": 0.8741333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0862/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": 103.3678970336914,
      "learning_rate": 6.331559480312315e-05,
      "loss": 3.2143,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6749333333333334,
      "eval_loss": 1.506548285484314,
      "eval_runtime": 7.4215,
      "eval_samples_per_second": 505.29,
      "eval_steps_per_second": 7.95,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 194.22702026367188,
      "learning_rate": 4.581559480312316e-05,
      "loss": 0.9049,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8424,
      "eval_loss": 0.5655111074447632,
      "eval_runtime": 7.3314,
      "eval_samples_per_second": 511.497,
      "eval_steps_per_second": 8.048,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 76.3280029296875,
      "learning_rate": 2.4184405196876842e-05,
      "loss": 0.506,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8605333333333334,
      "eval_loss": 0.46046122908592224,
      "eval_runtime": 7.4313,
      "eval_samples_per_second": 504.625,
      "eval_steps_per_second": 7.939,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 166.2055206298828,
      "learning_rate": 6.684405196876842e-06,
      "loss": 0.4085,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8653333333333333,
      "eval_loss": 0.4224889576435089,
      "eval_runtime": 7.3541,
      "eval_samples_per_second": 509.923,
      "eval_steps_per_second": 8.023,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 72.59909057617188,
      "learning_rate": 0.0,
      "loss": 0.3715,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8741333333333333,
      "eval_loss": 0.416171133518219,
      "eval_runtime": 7.3119,
      "eval_samples_per_second": 512.859,
      "eval_steps_per_second": 8.069,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.088214619776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}