trainer_state.json

{
  "best_metric": 0.8845333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0815/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 76.8904037475586,
      "learning_rate": 7e-05,
      "loss": 3.3686,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6357333333333334,
      "eval_loss": 1.786314606666565,
      "eval_runtime": 7.5871,
      "eval_samples_per_second": 494.261,
      "eval_steps_per_second": 7.776,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 167.5823516845703,
      "learning_rate": 7e-05,
      "loss": 1.0651,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8178666666666666,
      "eval_loss": 0.6647362112998962,
      "eval_runtime": 7.5329,
      "eval_samples_per_second": 497.816,
      "eval_steps_per_second": 7.832,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 151.5993194580078,
      "learning_rate": 7e-05,
      "loss": 0.6017,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.86,
      "eval_loss": 0.5001676678657532,
      "eval_runtime": 7.5217,
      "eval_samples_per_second": 498.559,
      "eval_steps_per_second": 7.844,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 90.41737365722656,
      "learning_rate": 7e-05,
      "loss": 0.4347,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8746666666666667,
      "eval_loss": 0.4333180785179138,
      "eval_runtime": 7.531,
      "eval_samples_per_second": 497.943,
      "eval_steps_per_second": 7.834,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 91.10928344726562,
      "learning_rate": 7e-05,
      "loss": 0.3442,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8845333333333333,
      "eval_loss": 0.4008885622024536,
      "eval_runtime": 7.456,
      "eval_samples_per_second": 502.949,
      "eval_steps_per_second": 7.913,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 4.088214619776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}