trainer_state.json

{
  "best_metric": 0.8722666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0697/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": 122.68234252929688,
      "learning_rate": 6.125e-05,
      "loss": 3.3329,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6432,
      "eval_loss": 2.095693588256836,
      "eval_runtime": 7.8567,
      "eval_samples_per_second": 477.301,
      "eval_steps_per_second": 7.51,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 132.23675537109375,
      "learning_rate": 5.2499999999999995e-05,
      "loss": 1.0986,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8034666666666667,
      "eval_loss": 0.6784162521362305,
      "eval_runtime": 7.6486,
      "eval_samples_per_second": 490.285,
      "eval_steps_per_second": 7.714,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 101.2192611694336,
      "learning_rate": 4.374999999999999e-05,
      "loss": 0.5824,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8373333333333334,
      "eval_loss": 0.530769407749176,
      "eval_runtime": 7.4921,
      "eval_samples_per_second": 500.528,
      "eval_steps_per_second": 7.875,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 108.42738342285156,
      "learning_rate": 3.5e-05,
      "loss": 0.45,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8568,
      "eval_loss": 0.459261029958725,
      "eval_runtime": 7.5596,
      "eval_samples_per_second": 496.056,
      "eval_steps_per_second": 7.805,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 381.1859436035156,
      "learning_rate": 2.6249999999999998e-05,
      "loss": 0.372,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8669333333333333,
      "eval_loss": 0.4261119067668915,
      "eval_runtime": 9.141,
      "eval_samples_per_second": 410.24,
      "eval_steps_per_second": 6.454,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 132.71316528320312,
      "learning_rate": 1.75e-05,
      "loss": 0.3185,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8722666666666666,
      "eval_loss": 0.40768924355506897,
      "eval_runtime": 7.5677,
      "eval_samples_per_second": 495.525,
      "eval_steps_per_second": 7.796,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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
}