时代的1/5训练似乎停滞不前，没有进展。

Question

我目前正在尝试训练我自己的图像识别模型，我有机会在Jason Brownlee教程上训练一个模型，我对训练超过一个班做了一些调整，当训练开始时我被困住了，因为第一个时代没有任何进展。下面是我的密码。我目前的数据集只有32张照片的汽车和自行车，所有的JPG格式，并已被注释。希望我能得到一些帮助，如果你需要更多的信息，我会很高兴提供它。

下面是从原来的

class ModelDataset(Dataset):
    # load the dataset definitions
    def load_dataset(self, dataset_dir, is_train=True):
        # define class
        lines = []
        with open('OpenLabeling-master/main/class_list.txt') as f:
            lines = f.readlines() 
        count = 1
        for line in lines:
            self.add_class("dataset", count, line)
            count += 1
        # define data locations
        images_dir = dataset_dir + '/images/'
        annotations_dir = dataset_dir + '/annots/'
        # find all images
        for filename in listdir(images_dir):
            # extract image id
            image_id = filename[:-4]
            # skip bad images
            if image_id in ['00090']:
                continue
            # skip all images after 150 if we are building the train set
            if is_train and int(image_id) >= 150:
                continue
            # skip all images before 150 if we are building the test/val set
            if not is_train and int(image_id) < 150:
                continue
            img_path = images_dir + filename
            ann_path = annotations_dir + image_id + '.xml'
            # add to dataset
            self.add_image('dataset', image_id=image_id, path=img_path, annotation=ann_path,class_ids=[0,1,2])
 
    # load all bounding boxes for an image
    def extract_boxes(self, filename):
        # load and parse the file
        root = ElementTree.parse(filename)
        boxes = list()
        # extract each bounding box
        box_class_list = list()
        for box in root.findall('.//bndbox'):
            xmin = int(box.find('xmin').text)
            ymin = int(box.find('ymin').text)
            xmax = int(box.find('xmax').text)
            ymax = int(box.find('ymax').text)
            coors = [xmin, ymin, xmax, ymax]
        # extract image dimensions
#       for name in root.findall('.//name'):
#           box_class_list.append(name.text)
        width = int(root.find('.//size/width').text)
        height = int(root.find('.//size/height').text)
        return boxes, width, height
 
    # load the masks for an image
    def load_mask(self, image_id):
        # get details of image
        info = self.image_info[image_id]
        # define box file location
        path = info['annotation']
        # load XML
        boxes, w, h = self.extract_boxes(path)
        # create one array for all masks, each on a different channel
        masks = zeros([h, w, len(boxes)], dtype='uint8')
        # create masks
        class_ids = list()
        for i in range(len(boxes)):
            box = boxes[i]
            print(box)
            row_s, row_e = box[1], box[3]
            col_s, col_e = box[0], box[2]
            if i == 0:
                masks[row_s:row_e, col_s:col_e, i] = 1
                class_ids.append(self.class_names.index('Car'))
            else:
                masks[row_s:row_e, col_s:col_e, i] = 2
                class_ids.append(self.class_names.index('Bicycle'))
        return masks, asarray(class_ids, dtype='int32')

我的配置如下

Configurations:
BACKBONE                       resnet101
BACKBONE_STRIDES               [4, 8, 16, 32, 64]
BATCH_SIZE                     1
BBOX_STD_DEV                   [0.1 0.1 0.2 0.2]
COMPUTE_BACKBONE_SHAPE         None
DETECTION_MAX_INSTANCES        100
DETECTION_MIN_CONFIDENCE       0.7
DETECTION_NMS_THRESHOLD        0.3
FPN_CLASSIF_FC_LAYERS_SIZE     1024
GPU_COUNT                      1
GRADIENT_CLIP_NORM             5.0
IMAGES_PER_GPU                 1
IMAGE_CHANNEL_COUNT            3
IMAGE_MAX_DIM                  1024
IMAGE_META_SIZE                15
IMAGE_MIN_DIM                  100
IMAGE_MIN_SCALE                0
IMAGE_RESIZE_MODE              square
IMAGE_SHAPE                    [1024 1024    3]
LEARNING_MOMENTUM              0.9
LEARNING_RATE                  0.001
LOSS_WEIGHTS                   {'rpn_class_loss': 1.0, 'rpn_bbox_loss': 1.0, 'mrcnn_class_loss': 1.0, 'mrcnn_bbox_loss': 1.0, 'mrcnn_mask_loss': 1.0}
MASK_POOL_SIZE                 14
MASK_SHAPE                     [28, 28]
MAX_GT_INSTANCES               100
MEAN_PIXEL                     [123.7 116.8 103.9]
MINI_MASK_SHAPE                (56, 56)
NAME                           trained_cfg
NUM_CLASSES                    3
POOL_SIZE                      7
POST_NMS_ROIS_INFERENCE        1000
POST_NMS_ROIS_TRAINING         2000
PRE_NMS_LIMIT                  6000
ROI_POSITIVE_RATIO             0.33
RPN_ANCHOR_RATIOS              [0.5, 1, 2]
RPN_ANCHOR_SCALES              (32, 64, 128, 256, 512)
RPN_ANCHOR_STRIDE              1
RPN_BBOX_STD_DEV               [0.1 0.1 0.2 0.2]
RPN_NMS_THRESHOLD              0.7
RPN_TRAIN_ANCHORS_PER_IMAGE    256
STEPS_PER_EPOCH                100
TOP_DOWN_PYRAMID_SIZE          256
TRAIN_BN                       False
TRAIN_ROIS_PER_IMAGE           100
USE_MINI_MASK                  True
USE_RPN_ROIS                   True
VALIDATION_STEPS               50
WEIGHT_DECAY                   0.0001

准备培训模型的代码

model = MaskRCNN(mode='training', model_dir='./', config=config)

model.load_weights('mask_rcnn_coco.h5', by_name=True, exclude=["mrcnn_class_logits", "mrcnn_bbox_fc",  "mrcnn_bbox", "mrcnn_mask"])

model.train(train_set, test_set, learning_rate=config.LEARNING_RATE, epochs=5, layers='heads')

Answer 1

我设法让它开始工作，这个时代开始运作了。问题在于所编辑的代码，我已经展示了我所做的代码更改的示例。它不知何故没有在extract_boxes方法中追加坐标，这会在print语句中造成无限循环。

    # load all bounding boxes for an image
    def extract_boxes(self, filename):
        # load and parse the file
        root = ElementTree.parse(filename)
        boxes = list()
        # extract each bounding box
        box_class_list = list()
        for box in root.findall('.//object'):
            name = box.find('name').text
            xmin = int(box.find('./bndbox/xmin').text)
            ymin = int(box.find('./bndbox/ymin').text)
            xmax = int(box.find('./bndbox/xmax').text)
            ymax = int(box.find('./bndbox/ymax').text)
            coors = [xmin, ymin, xmax, ymax, name]
            if name=='Car' or name=='Bicycle':
                boxes.append(coors)
        width = int(root.find('.//size/width').text)
        height = int(root.find('.//size/height').text)
        return boxes, width, height
 
    # load the masks for an image
    def load_mask(self, image_id):
        # get details of image
        y = 1
        info = self.image_info[image_id]
        # define box file location
        path = info['annotation']
#       print(path)
#       print("_____")
        # load XML
        boxes, w, h = self.extract_boxes(path)
#       print(boxes)
        # create one array for all masks, each on a different channel
        masks = zeros([h, w, len(boxes)], dtype='uint8')
        # create masks
        class_ids = list()
        for i in range(len(boxes)):
            box = boxes[i]
            row_s, row_e = box[1], box[3]
            col_s, col_e = box[0], box[2]
            if i == 0:
                masks[row_s:row_e, col_s:col_e, i] = 1
                class_ids.append(self.class_names.index('Car'))
            else:               
                masks[row_s:row_e, col_s:col_e, i] = 2
                class_ids.append(self.class_names.index('Bicycle'))
        return masks, asarray(class_ids, dtype='int32')