trainer_state.json

{
  "best_metric": 0.9064,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0472/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": 76.39278411865234,
      "learning_rate": 0.0001,
      "loss": 2.7844,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7424,
      "eval_loss": 0.9664188623428345,
      "eval_runtime": 7.9235,
      "eval_samples_per_second": 473.273,
      "eval_steps_per_second": 7.446,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 112.61178588867188,
      "learning_rate": 0.0001,
      "loss": 0.7297,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8536,
      "eval_loss": 0.489338755607605,
      "eval_runtime": 7.9567,
      "eval_samples_per_second": 471.3,
      "eval_steps_per_second": 7.415,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 113.12704467773438,
      "learning_rate": 0.0001,
      "loss": 0.4409,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8757333333333334,
      "eval_loss": 0.4061494767665863,
      "eval_runtime": 7.6802,
      "eval_samples_per_second": 488.266,
      "eval_steps_per_second": 7.682,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 114.21883392333984,
      "learning_rate": 0.0001,
      "loss": 0.3196,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8848,
      "eval_loss": 0.3671186864376068,
      "eval_runtime": 8.4207,
      "eval_samples_per_second": 445.333,
      "eval_steps_per_second": 7.007,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 67.83946228027344,
      "learning_rate": 0.0001,
      "loss": 0.2363,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8962666666666667,
      "eval_loss": 0.35183945298194885,
      "eval_runtime": 8.3511,
      "eval_samples_per_second": 449.044,
      "eval_steps_per_second": 7.065,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 107.41410827636719,
      "learning_rate": 0.0001,
      "loss": 0.1799,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9024,
      "eval_loss": 0.3330535590648651,
      "eval_runtime": 8.1149,
      "eval_samples_per_second": 462.114,
      "eval_steps_per_second": 7.271,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 80.69808197021484,
      "learning_rate": 0.0001,
      "loss": 0.1416,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9050666666666667,
      "eval_loss": 0.33134129643440247,
      "eval_runtime": 8.2635,
      "eval_samples_per_second": 453.803,
      "eval_steps_per_second": 7.14,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 68.13387298583984,
      "learning_rate": 0.0001,
      "loss": 0.1094,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9064,
      "eval_loss": 0.35234490036964417,
      "eval_runtime": 8.3164,
      "eval_samples_per_second": 450.915,
      "eval_steps_per_second": 7.094,
      "step": 2664
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}