trainer_state.json

{
  "best_metric": 0.8594666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0768/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": 42.018524169921875,
      "learning_rate": 0.0003,
      "loss": 1.6183,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8002666666666667,
      "eval_loss": 0.6766745448112488,
      "eval_runtime": 7.3988,
      "eval_samples_per_second": 506.836,
      "eval_steps_per_second": 7.974,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 45.84018325805664,
      "learning_rate": 0.0003,
      "loss": 0.5182,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8330666666666666,
      "eval_loss": 0.5119360089302063,
      "eval_runtime": 7.3804,
      "eval_samples_per_second": 508.102,
      "eval_steps_per_second": 7.994,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 33.7328987121582,
      "learning_rate": 0.0003,
      "loss": 0.3345,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8490666666666666,
      "eval_loss": 0.48920953273773193,
      "eval_runtime": 7.3627,
      "eval_samples_per_second": 509.322,
      "eval_steps_per_second": 8.013,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 35.122806549072266,
      "learning_rate": 0.0003,
      "loss": 0.2384,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8378666666666666,
      "eval_loss": 0.5706325769424438,
      "eval_runtime": 7.3806,
      "eval_samples_per_second": 508.09,
      "eval_steps_per_second": 7.994,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 37.13829803466797,
      "learning_rate": 0.0003,
      "loss": 0.1713,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8565333333333334,
      "eval_loss": 0.5373947620391846,
      "eval_runtime": 7.3772,
      "eval_samples_per_second": 508.322,
      "eval_steps_per_second": 7.998,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 25.979949951171875,
      "learning_rate": 0.0003,
      "loss": 0.1408,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8594666666666667,
      "eval_loss": 0.5519552230834961,
      "eval_runtime": 7.3886,
      "eval_samples_per_second": 507.541,
      "eval_steps_per_second": 7.985,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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
}