trainer_state.json

{
  "best_metric": 0.8696,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0956/checkpoints/checkpoint-2997",
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 2997,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 162.41836547851562,
      "learning_rate": 3e-05,
      "loss": 3.8473,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4066666666666667,
      "eval_loss": 3.709223985671997,
      "eval_runtime": 7.5629,
      "eval_samples_per_second": 495.844,
      "eval_steps_per_second": 7.801,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 140.19288635253906,
      "learning_rate": 3e-05,
      "loss": 3.0331,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6186666666666667,
      "eval_loss": 2.233522891998291,
      "eval_runtime": 7.7105,
      "eval_samples_per_second": 486.35,
      "eval_steps_per_second": 7.652,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 144.09326171875,
      "learning_rate": 3e-05,
      "loss": 1.4132,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.76,
      "eval_loss": 0.9559027552604675,
      "eval_runtime": 7.6855,
      "eval_samples_per_second": 487.931,
      "eval_steps_per_second": 7.677,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 245.33340454101562,
      "learning_rate": 3e-05,
      "loss": 0.8431,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8130666666666667,
      "eval_loss": 0.6744756698608398,
      "eval_runtime": 7.7989,
      "eval_samples_per_second": 480.838,
      "eval_steps_per_second": 7.565,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 180.73423767089844,
      "learning_rate": 3e-05,
      "loss": 0.6635,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8338666666666666,
      "eval_loss": 0.5643283724784851,
      "eval_runtime": 7.7122,
      "eval_samples_per_second": 486.24,
      "eval_steps_per_second": 7.65,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 150.23997497558594,
      "learning_rate": 3e-05,
      "loss": 0.5456,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8477333333333333,
      "eval_loss": 0.5088294744491577,
      "eval_runtime": 7.7255,
      "eval_samples_per_second": 485.407,
      "eval_steps_per_second": 7.637,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 469.0445556640625,
      "learning_rate": 3e-05,
      "loss": 0.4741,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8562666666666666,
      "eval_loss": 0.48997870087623596,
      "eval_runtime": 7.7446,
      "eval_samples_per_second": 484.209,
      "eval_steps_per_second": 7.618,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 95.25812530517578,
      "learning_rate": 3e-05,
      "loss": 0.4244,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8672,
      "eval_loss": 0.43941009044647217,
      "eval_runtime": 7.692,
      "eval_samples_per_second": 487.517,
      "eval_steps_per_second": 7.67,
      "step": 2664
    },
    {
      "epoch": 9.0,
      "grad_norm": 293.98577880859375,
      "learning_rate": 3e-05,
      "loss": 0.3857,
      "step": 2997
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8696,
      "eval_loss": 0.42392534017562866,
      "eval_runtime": 7.6973,
      "eval_samples_per_second": 487.181,
      "eval_steps_per_second": 7.665,
      "step": 2997
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7.3587863155968e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}