trainer_state.json

{
  "best_metric": 0.8826666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0841/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 27.02740478515625,
      "learning_rate": 0.0003,
      "loss": 1.5217,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8216,
      "eval_loss": 0.5826595425605774,
      "eval_runtime": 8.4238,
      "eval_samples_per_second": 445.166,
      "eval_steps_per_second": 7.004,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 29.30405044555664,
      "learning_rate": 0.0003,
      "loss": 0.4654,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8642666666666666,
      "eval_loss": 0.4705454409122467,
      "eval_runtime": 7.3382,
      "eval_samples_per_second": 511.026,
      "eval_steps_per_second": 8.04,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 29.838224411010742,
      "learning_rate": 0.0003,
      "loss": 0.2904,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8752,
      "eval_loss": 0.4420243501663208,
      "eval_runtime": 7.3967,
      "eval_samples_per_second": 506.984,
      "eval_steps_per_second": 7.977,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 70.53434753417969,
      "learning_rate": 0.0003,
      "loss": 0.1922,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8709333333333333,
      "eval_loss": 0.46980226039886475,
      "eval_runtime": 7.3867,
      "eval_samples_per_second": 507.67,
      "eval_steps_per_second": 7.987,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 26.233779907226562,
      "learning_rate": 0.0003,
      "loss": 0.1396,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8682666666666666,
      "eval_loss": 0.5048604607582092,
      "eval_runtime": 7.3542,
      "eval_samples_per_second": 509.916,
      "eval_steps_per_second": 8.023,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 29.690643310546875,
      "learning_rate": 0.0003,
      "loss": 0.1186,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8826666666666667,
      "eval_loss": 0.47286587953567505,
      "eval_runtime": 7.3606,
      "eval_samples_per_second": 509.47,
      "eval_steps_per_second": 8.016,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}