trainer_state.json

{
  "best_metric": 0.8701333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0439/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": 44.446510314941406,
      "learning_rate": 0.0005,
      "loss": 1.347,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7952,
      "eval_loss": 0.641343891620636,
      "eval_runtime": 15.5252,
      "eval_samples_per_second": 241.542,
      "eval_steps_per_second": 3.8,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 31.018463134765625,
      "learning_rate": 0.0005,
      "loss": 0.5121,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8402666666666667,
      "eval_loss": 0.5077934861183167,
      "eval_runtime": 15.5044,
      "eval_samples_per_second": 241.867,
      "eval_steps_per_second": 3.805,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 31.085983276367188,
      "learning_rate": 0.0005,
      "loss": 0.3585,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8453333333333334,
      "eval_loss": 0.5144662261009216,
      "eval_runtime": 15.5219,
      "eval_samples_per_second": 241.595,
      "eval_steps_per_second": 3.801,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 28.263837814331055,
      "learning_rate": 0.0005,
      "loss": 0.2658,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8434666666666667,
      "eval_loss": 0.5081769824028015,
      "eval_runtime": 15.5636,
      "eval_samples_per_second": 240.946,
      "eval_steps_per_second": 3.791,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 20.936574935913086,
      "learning_rate": 0.0005,
      "loss": 0.2154,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8541333333333333,
      "eval_loss": 0.5146492719650269,
      "eval_runtime": 15.5803,
      "eval_samples_per_second": 240.689,
      "eval_steps_per_second": 3.787,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 23.12115478515625,
      "learning_rate": 0.0005,
      "loss": 0.1754,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8589333333333333,
      "eval_loss": 0.49113398790359497,
      "eval_runtime": 15.5748,
      "eval_samples_per_second": 240.774,
      "eval_steps_per_second": 3.788,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 22.891674041748047,
      "learning_rate": 0.0005,
      "loss": 0.1401,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8557333333333333,
      "eval_loss": 0.5404118299484253,
      "eval_runtime": 16.1147,
      "eval_samples_per_second": 232.706,
      "eval_steps_per_second": 3.661,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 28.590261459350586,
      "learning_rate": 0.0005,
      "loss": 0.1308,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8701333333333333,
      "eval_loss": 0.5056899189949036,
      "eval_runtime": 15.7916,
      "eval_samples_per_second": 237.469,
      "eval_steps_per_second": 3.736,
      "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
}