trainer_state.json

{
  "best_metric": 0.8586666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0646/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": 138.0891571044922,
      "learning_rate": 6.653391037658466e-05,
      "loss": 3.2734,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5962666666666666,
      "eval_loss": 1.8463690280914307,
      "eval_runtime": 13.9025,
      "eval_samples_per_second": 269.736,
      "eval_steps_per_second": 4.244,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 389.82464599609375,
      "learning_rate": 5.682214306505567e-05,
      "loss": 1.0828,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7954666666666667,
      "eval_loss": 0.7384634017944336,
      "eval_runtime": 13.8183,
      "eval_samples_per_second": 271.38,
      "eval_steps_per_second": 4.27,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 82.43489074707031,
      "learning_rate": 4.2788232688471e-05,
      "loss": 0.6112,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8357333333333333,
      "eval_loss": 0.5595542788505554,
      "eval_runtime": 13.7987,
      "eval_samples_per_second": 271.764,
      "eval_steps_per_second": 4.276,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 171.4561309814453,
      "learning_rate": 2.7211767311528996e-05,
      "loss": 0.4413,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8437333333333333,
      "eval_loss": 0.5136469006538391,
      "eval_runtime": 13.5783,
      "eval_samples_per_second": 276.175,
      "eval_steps_per_second": 4.345,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 164.77854919433594,
      "learning_rate": 1.3177856934944328e-05,
      "loss": 0.3459,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8586666666666667,
      "eval_loss": 0.4859790802001953,
      "eval_runtime": 13.4827,
      "eval_samples_per_second": 278.135,
      "eval_steps_per_second": 4.376,
      "step": 1665
    }
  ],
  "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": 4.088214619776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}