trainer_state.json

{
  "best_metric": 0.8877333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0830/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": 97.32987976074219,
      "learning_rate": 0.0001,
      "loss": 2.7963,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7104,
      "eval_loss": 1.0868257284164429,
      "eval_runtime": 8.8017,
      "eval_samples_per_second": 426.055,
      "eval_steps_per_second": 6.703,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 164.62606811523438,
      "learning_rate": 0.0001,
      "loss": 0.7691,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8394666666666667,
      "eval_loss": 0.5487115383148193,
      "eval_runtime": 7.6388,
      "eval_samples_per_second": 490.914,
      "eval_steps_per_second": 7.724,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 57.00876235961914,
      "learning_rate": 0.0001,
      "loss": 0.4623,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8554666666666667,
      "eval_loss": 0.4586714208126068,
      "eval_runtime": 7.5562,
      "eval_samples_per_second": 496.282,
      "eval_steps_per_second": 7.808,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 114.31270599365234,
      "learning_rate": 0.0001,
      "loss": 0.3346,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8752,
      "eval_loss": 0.40806737542152405,
      "eval_runtime": 7.6919,
      "eval_samples_per_second": 487.528,
      "eval_steps_per_second": 7.67,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 74.4852523803711,
      "learning_rate": 0.0001,
      "loss": 0.2519,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8877333333333334,
      "eval_loss": 0.36694660782814026,
      "eval_runtime": 7.5601,
      "eval_samples_per_second": 496.022,
      "eval_steps_per_second": 7.804,
      "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
}