训练 detectron2 遇到的 bug : 训练过程中无法评估测试集

写在前面的话

最近在使用 detectron2 训练网络的时候，我想在训练的过程中能看到模型对测试集的检测效果，我在yaml配置文件设置的是5000次迭代输出一次对测试集的评估结果，设置代码如下：

cfg.TEST.EVAL_PERIOD = 5000 #打印输出

但是当我训练到第5000个迭代的时候却没有如我设想的出现测试集的评估结果，显示的是一个 warning ，报错如下：

No evaluator found. Use DefaultTrainer.test(evaluators=), or implement its build_evaluator method.

解决方法

这个报错（warning）的出处是 detectron2/engine/defaults.py 里面官方原始build_evaluator函数

官方原始build_evaluator函数

    @classmethod
    def build_evaluator(cls, cfg, dataset_name):
        """
        Returns:
            DatasetEvaluator or None

        It is not implemented by default.
        """
        raise NotImplementedError(
            """
If you want DefaultTrainer to automatically run evaluation,
please implement `build_evaluator()` in subclasses (see train_net.py for example).
Alternatively, you can call evaluation functions yourself (see Colab balloon tutorial for example).
"""
        )

github上面的对这个issue的讨论：
https://github.com/facebookresearch/detectron2/issues/951

参考github上面的答案：
See examples https://github.com/facebookresearch/detectron2/blob/master/tools/train_net.py

官方源码的主目录下有一个 train_net.py 文件，里面也有build_evaluator函数，github上面的答案的意思是参考 train_net.py 里面 build_evaluator函数的内容．

我的方法比较简单粗暴，我也懒得参考，直接把原始的build_evaluator函数删除，然后把 train_net.py 里面 build_evaluator函数复制粘帖到原始的位置

修改后的build_evaluator函数（跟train_net.py 里面 build_evaluator函数一样）

    @classmethod
    def build_evaluator(cls, cfg, dataset_name, output_folder=None):
        """
        Create evaluator(s) for a given dataset.
        This uses the special metadata "evaluator_type" associated with each builtin dataset.
        For your own dataset, you can simply create an evaluator manually in your
        script and do not have to worry about the hacky if-else logic here.
        """
        if output_folder is None:
            output_folder = os.path.join(cfg.OUTPUT_DIR, "inference")
        evaluator_list = []
        evaluator_type = MetadataCatalog.get(dataset_name).evaluator_type
        if evaluator_type in ["sem_seg", "coco_panoptic_seg"]:
            evaluator_list.append(
                SemSegEvaluator(
                    dataset_name,
                    distributed=True,
                    num_classes=cfg.MODEL.SEM_SEG_HEAD.NUM_CLASSES,
                    ignore_label=cfg.MODEL.SEM_SEG_HEAD.IGNORE_VALUE,
                    output_dir=output_folder,
                )
            )
        if evaluator_type in ["coco", "coco_panoptic_seg"]:
            evaluator_list.append(COCOEvaluator(dataset_name, cfg, True, output_folder))
        if evaluator_type == "coco_panoptic_seg":
            evaluator_list.append(COCOPanopticEvaluator(dataset_name, output_folder))
        if evaluator_type == "cityscapes_instance":
            assert (
                torch.cuda.device_count() >= comm.get_rank()
            ), "CityscapesEvaluator currently do not work with multiple machines."
            return CityscapesInstanceEvaluator(dataset_name)
        if evaluator_type == "cityscapes_sem_seg":
            assert (
                torch.cuda.device_count() >= comm.get_rank()
            ), "CityscapesEvaluator currently do not work with multiple machines."
            return CityscapesSemSegEvaluator(dataset_name)
        elif evaluator_type == "pascal_voc":
            return PascalVOCDetectionEvaluator(dataset_name)
        elif evaluator_type == "lvis":
            return LVISEvaluator(dataset_name, cfg, True, output_folder)
        if len(evaluator_list) == 0:
            raise NotImplementedError(
                "no Evaluator for the dataset {} with the type {}".format(
                    dataset_name, evaluator_type
                )
            )
        elif len(evaluator_list) == 1:
            return evaluator_list[0]
        return DatasetEvaluators(evaluator_list)

注意：一些调用的函数需要 import 进来

'''记得加载函数！！！'''
from detectron2.evaluation import (
    CityscapesInstanceEvaluator,
    CityscapesSemSegEvaluator,
    COCOEvaluator,
    COCOPanopticEvaluator,
    DatasetEvaluators,
    LVISEvaluator,
    PascalVOCDetectionEvaluator,
    SemSegEvaluator,
    verify_results,
)

修改完成后，遇到的问题就解决了！！！