trainer_state.json

{
  "best_metric": 0.8594666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0185/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": 52.08506774902344,
      "learning_rate": 0.0005,
      "loss": 1.4068,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.792,
      "eval_loss": 0.7121949195861816,
      "eval_runtime": 13.4166,
      "eval_samples_per_second": 279.504,
      "eval_steps_per_second": 4.398,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 39.388397216796875,
      "learning_rate": 0.0005,
      "loss": 0.5169,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.832,
      "eval_loss": 0.5281139612197876,
      "eval_runtime": 13.2973,
      "eval_samples_per_second": 282.012,
      "eval_steps_per_second": 4.437,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 23.90403175354004,
      "learning_rate": 0.0005,
      "loss": 0.3557,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8512,
      "eval_loss": 0.5057083368301392,
      "eval_runtime": 13.2329,
      "eval_samples_per_second": 283.385,
      "eval_steps_per_second": 4.459,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 20.769189834594727,
      "learning_rate": 0.0005,
      "loss": 0.2749,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8533333333333334,
      "eval_loss": 0.503274142742157,
      "eval_runtime": 13.2216,
      "eval_samples_per_second": 283.627,
      "eval_steps_per_second": 4.462,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 27.287212371826172,
      "learning_rate": 0.0005,
      "loss": 0.2077,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8594666666666667,
      "eval_loss": 0.5162767767906189,
      "eval_runtime": 14.4618,
      "eval_samples_per_second": 259.303,
      "eval_steps_per_second": 4.08,
      "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
}