trainer_state.json

{
  "best_metric": 0.7944,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0986/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": 180.14109802246094,
      "learning_rate": 4e-05,
      "loss": 3.7601,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.448,
      "eval_loss": 3.400230884552002,
      "eval_runtime": 8.2763,
      "eval_samples_per_second": 453.103,
      "eval_steps_per_second": 7.129,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 211.23538208007812,
      "learning_rate": 3e-05,
      "loss": 2.3645,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6688,
      "eval_loss": 1.5139724016189575,
      "eval_runtime": 8.4725,
      "eval_samples_per_second": 442.608,
      "eval_steps_per_second": 6.964,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 152.18666076660156,
      "learning_rate": 2e-05,
      "loss": 1.1792,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7642666666666666,
      "eval_loss": 0.9442797303199768,
      "eval_runtime": 8.4465,
      "eval_samples_per_second": 443.973,
      "eval_steps_per_second": 6.985,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 481.51190185546875,
      "learning_rate": 1e-05,
      "loss": 0.8939,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7845333333333333,
      "eval_loss": 0.7886518239974976,
      "eval_runtime": 8.2894,
      "eval_samples_per_second": 452.382,
      "eval_steps_per_second": 7.117,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 180.92123413085938,
      "learning_rate": 0.0,
      "loss": 0.7877,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7944,
      "eval_loss": 0.7550519704818726,
      "eval_runtime": 8.1824,
      "eval_samples_per_second": 458.298,
      "eval_steps_per_second": 7.211,
      "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
}