在生成YML文件时，时不时出现Empty trainin g data was given.

Question

在生成yml文件时，时不时会出现以下报错

D:\miniconda\envs\py37\python.exe "C:/Users/38273/Desktop/code/hope/whole test/facetest/13.py"
images/cap20230403161937
id: 1
face: ()
Traceback (most recent call last):
  File "C:/Users/38273/Desktop/code/hope/whole test/facetest/13.py", line 125, in train_face
    recognizer.train(faces, np.array(ids))
cv2.error: OpenCV(4.7.0) D:\a\opencv-python\opencv-python\opencv_contrib\modules\face\src\lbph_faces.cpp:362: error: (-210:Unsupported format or combination of formats) Empty trainin
g data was given. You'll need more than one sample to learn a model. in function 'cv::face::LBPH::train'

进程已结束，退出代码为 -1073740791 (0xC0000409)

以下是代码

    def save_face(self):
        self.open_camera()
        path = './photo'
        path1 = './face_id'
        exist = os.path.exists(path)
        exist1 = os.path.exists(path1)
        if not exist:
            os.mkdir(path)
        if not exist1:
            os.mkdir(path1)

        id = '1'
        name = '张三'
        classroom = '通信1902'
        name = self.pinyin(name)
        classroom = self.pinyin(classroom)
        # cap = cv2.VideoCapture(0)
        # ret, frame = cap.read()
        if self.cap.isOpened():
            FName = fr"images/cap{time.strftime('%Y%m%d%H%M%S', time.localtime())}"
            print(FName)
            self.showImage.save('./photo/' + id + '.' + name + '.' + classroom + '.jpg')
            # cv2.imwrite(path + str(id) + '.' + str(classroom) + str(name) + ".jpg", frame)
        else:
            QMessageBox.critical(self, '错误', '摄像头未打开！')
            return None

    def get_face_id(self):
        self.save_face()
        path = './photo'
        picture = []
        ids = []
        imagepaths = [os.path.join(path, f) for f in os.listdir(path)]
        detector = cv2.CascadeClassifier('D:/opencv/opencv/sources/data/haarcascades/haarcascade_frontalface_alt2.xml')

        for imagepath in imagepaths:
            img = Image.open(imagepath).convert('L')
            img_np = np.array(img, 'uint8')
            faces = detector.detectMultiScale(img_np)
            id = int(os.path.split(imagepath)[1].split('.')[0])
            for x, y, w, h in faces:
                ids.append(id)
                picture.append(img_np[y:y + h, x:x + w])
        print('id:', id)
        print('face:', faces)
        return picture, ids
        #
        # recognizer = cv2.face.LBPHFaceRecognizer_create()
        # recognizer.train(picture, np.array(ids))
        # recognizer.write('./face_id/face_id.yml')

    def train_face(self):
        # self.open_camera
        faces, ids = self.get_face_id()
        recognizer = cv2.face.LBPHFaceRecognizer_create()
        recognizer.train(faces, np.array(ids))
        recognizer.write('./face_id/face_id.yml')
        return QMessageBox.information(
            self,
            '成功',
            '保存成功'
        )

Answer 1

"Empty training data was given. You'll need more than one sample to learn a model." 表示训练数据为空
可能原因：
训练数据集太小。确保你有足够数量的图像来训练模型。
数据集中没有正确的标签。确保所有人脸图像都有正确的标签和ID，以便模型能够识别它们。
训练数据集的路径有误。确保你提供了正确的图像路径，并且程序可以正确地读取它们。
图像文件格式有误。确保图像文件格式是支持的，并且它们可以被读取和解码。