trainer_state.json

{
  "best_metric": 0.8666666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0677/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": 262.9297180175781,
      "learning_rate": 6.653391037658466e-05,
      "loss": 3.4507,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5552,
      "eval_loss": 2.280982255935669,
      "eval_runtime": 15.4358,
      "eval_samples_per_second": 242.942,
      "eval_steps_per_second": 3.822,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 65.12991333007812,
      "learning_rate": 5.682214306505567e-05,
      "loss": 1.2729,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7754666666666666,
      "eval_loss": 0.8016969561576843,
      "eval_runtime": 15.422,
      "eval_samples_per_second": 243.159,
      "eval_steps_per_second": 3.826,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 121.97489929199219,
      "learning_rate": 4.2788232688471e-05,
      "loss": 0.6911,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8226666666666667,
      "eval_loss": 0.6060190796852112,
      "eval_runtime": 15.7512,
      "eval_samples_per_second": 238.078,
      "eval_steps_per_second": 3.746,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 133.89389038085938,
      "learning_rate": 2.7211767311528996e-05,
      "loss": 0.5297,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8514666666666667,
      "eval_loss": 0.5109049081802368,
      "eval_runtime": 15.4212,
      "eval_samples_per_second": 243.172,
      "eval_steps_per_second": 3.826,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 161.7786407470703,
      "learning_rate": 1.3177856934944328e-05,
      "loss": 0.4421,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8666666666666667,
      "eval_loss": 0.457114577293396,
      "eval_runtime": 14.9495,
      "eval_samples_per_second": 250.844,
      "eval_steps_per_second": 3.947,
      "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
}