trainer_state.json

{
  "best_metric": 0.8728,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0038/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": 78.6984634399414,
      "learning_rate": 8.728616793536588e-05,
      "loss": 3.1043,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6490666666666667,
      "eval_loss": 1.4619672298431396,
      "eval_runtime": 13.607,
      "eval_samples_per_second": 275.594,
      "eval_steps_per_second": 4.336,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 110.86795043945312,
      "learning_rate": 7.947199994035401e-05,
      "loss": 0.9625,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8056,
      "eval_loss": 0.6877768039703369,
      "eval_runtime": 13.1295,
      "eval_samples_per_second": 285.617,
      "eval_steps_per_second": 4.494,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 175.60641479492188,
      "learning_rate": 6.75e-05,
      "loss": 0.5881,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8496,
      "eval_loss": 0.5141373872756958,
      "eval_runtime": 13.2595,
      "eval_samples_per_second": 282.817,
      "eval_steps_per_second": 4.45,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 64.1792221069336,
      "learning_rate": 5.2814167995011876e-05,
      "loss": 0.4323,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8570666666666666,
      "eval_loss": 0.490458607673645,
      "eval_runtime": 13.0248,
      "eval_samples_per_second": 287.912,
      "eval_steps_per_second": 4.53,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 166.4523162841797,
      "learning_rate": 3.7185832004988136e-05,
      "loss": 0.3529,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8634666666666667,
      "eval_loss": 0.47059518098831177,
      "eval_runtime": 13.1414,
      "eval_samples_per_second": 285.357,
      "eval_steps_per_second": 4.49,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 161.69883728027344,
      "learning_rate": 2.250000000000001e-05,
      "loss": 0.2994,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8728,
      "eval_loss": 0.44414907693862915,
      "eval_runtime": 13.0441,
      "eval_samples_per_second": 287.486,
      "eval_steps_per_second": 4.523,
      "step": 1998
    }
  ],
  "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": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}