trainer_state.json

{
  "best_metric": 0.8869333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0509/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": 210.6145782470703,
      "learning_rate": 0.0001,
      "loss": 2.7962,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7330666666666666,
      "eval_loss": 1.0217713117599487,
      "eval_runtime": 13.1604,
      "eval_samples_per_second": 284.945,
      "eval_steps_per_second": 4.483,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 152.42138671875,
      "learning_rate": 0.0001,
      "loss": 0.7249,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8336,
      "eval_loss": 0.5179799795150757,
      "eval_runtime": 13.0028,
      "eval_samples_per_second": 288.4,
      "eval_steps_per_second": 4.537,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 135.20501708984375,
      "learning_rate": 0.0001,
      "loss": 0.4392,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8666666666666667,
      "eval_loss": 0.41690367460250854,
      "eval_runtime": 13.1848,
      "eval_samples_per_second": 284.419,
      "eval_steps_per_second": 4.475,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 83.49943542480469,
      "learning_rate": 0.0001,
      "loss": 0.3174,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8824,
      "eval_loss": 0.37888240814208984,
      "eval_runtime": 12.9861,
      "eval_samples_per_second": 288.771,
      "eval_steps_per_second": 4.543,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 67.60909271240234,
      "learning_rate": 0.0001,
      "loss": 0.2443,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8869333333333334,
      "eval_loss": 0.3672392666339874,
      "eval_runtime": 13.0531,
      "eval_samples_per_second": 287.289,
      "eval_steps_per_second": 4.52,
      "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
}