trainer_state.json

{
  "best_metric": 0.9013333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0083/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": 64.37771606445312,
      "learning_rate": 8.728616793536588e-05,
      "loss": 2.7787,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7696,
      "eval_loss": 0.894626259803772,
      "eval_runtime": 7.5311,
      "eval_samples_per_second": 497.938,
      "eval_steps_per_second": 7.834,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 76.82396697998047,
      "learning_rate": 7.947199994035401e-05,
      "loss": 0.6383,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8562666666666666,
      "eval_loss": 0.46326908469200134,
      "eval_runtime": 8.047,
      "eval_samples_per_second": 466.014,
      "eval_steps_per_second": 7.332,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 32.40639877319336,
      "learning_rate": 6.75e-05,
      "loss": 0.3946,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8837333333333334,
      "eval_loss": 0.3748651146888733,
      "eval_runtime": 8.1734,
      "eval_samples_per_second": 458.803,
      "eval_steps_per_second": 7.219,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 65.1263656616211,
      "learning_rate": 5.2814167995011876e-05,
      "loss": 0.2793,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.888,
      "eval_loss": 0.36405929923057556,
      "eval_runtime": 7.9457,
      "eval_samples_per_second": 471.955,
      "eval_steps_per_second": 7.425,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 240.65301513671875,
      "learning_rate": 3.7185832004988136e-05,
      "loss": 0.214,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8930666666666667,
      "eval_loss": 0.3450826108455658,
      "eval_runtime": 7.9159,
      "eval_samples_per_second": 473.73,
      "eval_steps_per_second": 7.453,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 97.2986831665039,
      "learning_rate": 2.250000000000001e-05,
      "loss": 0.1683,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9013333333333333,
      "eval_loss": 0.3219829499721527,
      "eval_runtime": 7.9685,
      "eval_samples_per_second": 470.603,
      "eval_steps_per_second": 7.404,
      "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
}