trainer_state.json

{
  "best_metric": 0.8786666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0416/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": 129.13365173339844,
      "learning_rate": 9e-05,
      "loss": 3.0266,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6832,
      "eval_loss": 1.2681673765182495,
      "eval_runtime": 8.8982,
      "eval_samples_per_second": 421.432,
      "eval_steps_per_second": 6.631,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 64.96981811523438,
      "learning_rate": 9e-05,
      "loss": 0.8476,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8229333333333333,
      "eval_loss": 0.6070020794868469,
      "eval_runtime": 7.8312,
      "eval_samples_per_second": 478.857,
      "eval_steps_per_second": 7.534,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 48.27806854248047,
      "learning_rate": 9e-05,
      "loss": 0.5359,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8554666666666667,
      "eval_loss": 0.4813118875026703,
      "eval_runtime": 7.8262,
      "eval_samples_per_second": 479.159,
      "eval_steps_per_second": 7.539,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 59.359535217285156,
      "learning_rate": 9e-05,
      "loss": 0.3976,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8656,
      "eval_loss": 0.4365074634552002,
      "eval_runtime": 7.9,
      "eval_samples_per_second": 474.681,
      "eval_steps_per_second": 7.468,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 98.37922668457031,
      "learning_rate": 9e-05,
      "loss": 0.3138,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8786666666666667,
      "eval_loss": 0.4090752601623535,
      "eval_runtime": 7.8582,
      "eval_samples_per_second": 477.211,
      "eval_steps_per_second": 7.508,
      "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
}