trainer_state.json

{
  "best_metric": 0.8776,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0028/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 144.87977600097656,
      "learning_rate": 3e-05,
      "loss": 3.8568,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.332,
      "eval_loss": 3.742933750152588,
      "eval_runtime": 9.5193,
      "eval_samples_per_second": 393.936,
      "eval_steps_per_second": 6.198,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 201.9223175048828,
      "learning_rate": 3e-05,
      "loss": 3.1957,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6016,
      "eval_loss": 2.4136393070220947,
      "eval_runtime": 9.3755,
      "eval_samples_per_second": 399.977,
      "eval_steps_per_second": 6.293,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 455.1341247558594,
      "learning_rate": 3e-05,
      "loss": 1.6472,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7557333333333334,
      "eval_loss": 1.0709738731384277,
      "eval_runtime": 9.4062,
      "eval_samples_per_second": 398.673,
      "eval_steps_per_second": 6.272,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 155.80934143066406,
      "learning_rate": 3e-05,
      "loss": 0.9005,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8077333333333333,
      "eval_loss": 0.7046900987625122,
      "eval_runtime": 9.4812,
      "eval_samples_per_second": 395.52,
      "eval_steps_per_second": 6.223,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 181.86093139648438,
      "learning_rate": 3e-05,
      "loss": 0.6609,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8376,
      "eval_loss": 0.5779275298118591,
      "eval_runtime": 9.5587,
      "eval_samples_per_second": 392.313,
      "eval_steps_per_second": 6.172,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 470.8018493652344,
      "learning_rate": 3e-05,
      "loss": 0.5369,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8573333333333333,
      "eval_loss": 0.501679539680481,
      "eval_runtime": 9.8911,
      "eval_samples_per_second": 379.128,
      "eval_steps_per_second": 5.965,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 217.91212463378906,
      "learning_rate": 3e-05,
      "loss": 0.4597,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8672,
      "eval_loss": 0.45517298579216003,
      "eval_runtime": 10.5301,
      "eval_samples_per_second": 356.121,
      "eval_steps_per_second": 5.603,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 184.84750366210938,
      "learning_rate": 3e-05,
      "loss": 0.4123,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8776,
      "eval_loss": 0.42092105746269226,
      "eval_runtime": 9.3447,
      "eval_samples_per_second": 401.296,
      "eval_steps_per_second": 6.314,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}