pybaseutils 2.0.8__tar.gz → 2.0.9__tar.gz

{pybaseutils-2.0.8 → pybaseutils-2.0.9}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pybaseutils
-Version: 2.0.8
+Version: 2.0.9
 Summary: pybaseutils
 Home-page: https://github.com/PanJinquan/base-utils
 Author: PanJinquan

{pybaseutils-2.0.8 → pybaseutils-2.0.9}/pybaseutils/__init__.py

@@ -5,4 +5,4 @@
     @Date   : 2019-05-07 17:40:27
 """
 
-__version__ = '2.0.8'
+__version__ = '2.0.9'

{pybaseutils-2.0.8 → pybaseutils-2.0.9}/pybaseutils/coords_utils.py

@@ -198,11 +198,12 @@ def shrink_polygon_pyclipper(polygon, ratio):
     return shrinked
 
 
-def get_square_boxes(boxes, use_max=True, baseline=-1):
+def get_square_boxes(boxes, use_max=True, use_mean=False, baseline=-1):
     """
     将boxes转换为正方形的boxes
     :param boxes:
-    :param use_max: 是否按照每个box(w,h)最大值进行转换
+    :param use_max: 是否按照每个box(w,h)最大值(True)/最小值(False)进行转换(默认)
+    :param use_mean: 是否按照每个box(w,h)平均值进行转换(优先级比use_mean高)
     :param baseline: 当baseline>0，表示正方形最小边长
     :return:
     """
@@ -213,6 +214,8 @@ def get_square_boxes(boxes, use_max=True, baseline=-1):
         b = np.max(center[:, 2:4], axis=1)
     else:
         b = np.min(center[:, 2:4], axis=1)
+    if use_mean:
+        b = np.mean(center[:, 2:4], axis=1)
     if baseline > 0:
         index = b < baseline
         b[index] = baseline

{pybaseutils-2.0.8 → pybaseutils-2.0.9}/pybaseutils/dataloader/parser_image_folder.py

@@ -39,7 +39,7 @@ class FolderDataset(parser_image_text.TextDataset):
     def __getitem__(self, index):
         """
         :param index:
-        :return: image,label
+        :return: {"image": image, "label": label}
         """
         item = self.item_list[index]
         bbox = item[2:] if len(item) == 6 else []
@@ -65,7 +65,7 @@ class FolderDataset(parser_image_text.TextDataset):
         item_list = []
         for i, dir in enumerate(filename):
             print("loading image from:{}".format(dir))
-            if not os.path.exists(dir): raise Exception("image_dir:{}".format(dir))
+            if not os.path.exists(dir): raise Exception("文件不存在，image_dir:{}".format(dir))
             paths, labels = file_utils.get_files_labels(dir, postfix=file_utils.IMG_POSTFIX)
             # TODO # 避免多个数据集的相同的label
             if use_sub:  labels = [os.path.join(str(i), l) for l in labels]
@@ -79,6 +79,7 @@ if __name__ == '__main__':
     from torchvision import transforms
 
     image_dir = ['/home/PKing/nasdata/release/infrastructure/DMClassification/data/dataset/train']
+    class_name = None
     input_size = [224, 224]
     rgb_mean = [0., 0., 0.]
     rgb_std = [1.0, 1.0, 1.0]
@@ -90,7 +91,7 @@ if __name__ == '__main__':
     dataset = FolderDataset(image_dir=image_dir,
                             transform=transform,
                             shuffle=True,
-                            class_name=None,
+                            class_name=class_name,
                             resample=True,
                             disp=True)
     for i in range(len(dataset)):

{pybaseutils-2.0.8 → pybaseutils-2.0.9}/pybaseutils/dataloader/parser_image_text.py

@@ -35,8 +35,9 @@ class TextDataset(Dataset):
         :param phase:
         :param disp:
         :param check:
-        :param kwargs:
+        :param kwargs: crop_scale,resample,save_info
         """
+        self.kwargs = kwargs
         self.use_rgb = use_rgb
         self.transform = transform
         self.phase = phase
@@ -55,9 +56,9 @@ class TextDataset(Dataset):
         self.num_images = len(self.item_list)
         self.classes = list(self.class_dict.values())
         self.num_classes = max(self.classes) + 1
-        self.info(save=kwargs.get("save", ""))
+        self.info(save_info=kwargs.get("save_info", ""))
 
-    def info(self, save=""):
+    def info(self, save_info=""):
         print("----------------------- {} DATASET INFO -----------------------".format(self.phase.upper()))
         print("Dataset have images   :{}".format(len(self.item_list)))
         print("Dataset class_name    :{}".format(self.class_name))
@@ -66,17 +67,17 @@ class TextDataset(Dataset):
         print("Dataset num images    :{}".format(len(self.item_list)))
         print("Dataset num_classes   :{}".format(self.num_classes))
         print("Dataset resample      :{}".format(self.resample))
-        if save:
-            if not os.path.exists(save): os.makedirs(save)
+        if save_info:
+            if not os.path.exists(save_info): os.makedirs(save_info)
             m = np.mean(list(self.class_count.values()))
             class_lack = {n: c for n, c in self.class_count.items() if c < m * 0.5}
             class_lack = sorted(class_lack.items(), key=lambda x: x[1], reverse=True)
             class_lack = {n[0]: n[1] for n in class_lack}
             class_lack.update({"mean": m})
-            file_utils.write_json_path(os.path.join(save, f"{self.phase}_class_dict.json"), self.class_dict)
-            file_utils.write_json_path(os.path.join(save, f"{self.phase}_class_count.json"), self.class_count)
-            file_utils.write_list_data(os.path.join(save, f"{self.phase}_class_name.txt"), self.class_name)
-            file_utils.write_json_path(os.path.join(save, f"{self.phase}_class_lack.json"), class_lack)
+            file_utils.write_json_path(os.path.join(save_info, f"{self.phase}_class_dict.json"), self.class_dict)
+            file_utils.write_json_path(os.path.join(save_info, f"{self.phase}_class_count.json"), self.class_count)
+            file_utils.write_list_data(os.path.join(save_info, f"{self.phase}_class_name.txt"), self.class_name)
+            file_utils.write_json_path(os.path.join(save_info, f"{self.phase}_class_lack.json"), class_lack)
             print("loss_labels: {}".format(class_lack))
         print("------------------------------------------------------------------")
 
@@ -91,8 +92,8 @@ class TextDataset(Dataset):
         """
         data_list = self.load_dataset(filename, data_root=data_root)
         if not self.class_name:
-            class_name = list(set([d[1] for d in data_list]))
-            self.class_name, self.class_dict = self.parser_classes(class_name)
+            self.class_name = list(set([d[1] for d in data_list]))
+            self.class_name, self.class_dict = self.parser_classes(self.class_name)
         item_list = []
         for data in data_list:
             label = data[1]
@@ -100,6 +101,8 @@ class TextDataset(Dataset):
             data[1] = self.class_dict[label]
             item_list.append(data)
         if check: item_list = self.check_item(item_list)
+        assert self.class_name, f"类别为空，请检查，class_name={self.class_name}"
+        assert item_list, f"文件列表为空，请检查输入数据，filename={filename}"
         if shuffle:
             random.seed(100)
             random.shuffle(item_list)
@@ -156,13 +159,13 @@ class TextDataset(Dataset):
     def __getitem__(self, index):
         """
         :param index:
-        :return: image,label
+        :return: {"image": image, "label": label}
         """
         item = self.item_list[index]
         bbox = item[2:] if len(item) == 6 else []
         image_file, label = item[0], item[1]
         image = self.read_image(image_file, use_rgb=self.use_rgb)
-        image = self.crop_image(image, bbox=bbox) if bbox else image
+        image = self.crop_image(image, bbox=bbox, scale=self.kwargs.get("crop_scale", [1.2, 1.2])) if bbox else image
         if self.transform:
             image = Image.fromarray(image)
             image = self.transform(image)
@@ -176,10 +179,11 @@ class TextDataset(Dataset):
             self.item_list = self.data_resample.update(True)
         return len(self.item_list)
 
-    def crop_image(self, image, bbox):
-        """采集图片"""
+    def crop_image(self, image, bbox, scale=[1.2, 1.2]):
+        """裁剪图片"""
         if not bbox: return image
-        boxes = image_utils.extend_xyxy([bbox], scale=[1.2, 1.2])
+        boxes = image_utils.get_square_boxes([bbox], use_max=True)
+        boxes = image_utils.extend_xyxy(boxes, scale=scale)
         image = image_utils.get_boxes_crop(image, boxes)[0]
         return image
 
@@ -241,10 +245,10 @@ if __name__ == '__main__':
     from torchvision import transforms
 
     filenames = [
-        # '/home/PKing/nasdata/tmp/tmp/RealFakeFace/anti-spoofing-images-v1/CelebA-Spoof/CelebA_Spoof/test.txt',
-        '/home/PKing/nasdata/tmp/tmp/RealFakeFace/anti-spoofing-images-v1/CelebA-Spoof/CelebA_Spoof/Data/train.txt',
+        '/home/PKing/nasdata/tmp/tmp/RealFakeFace/anti-spoofing-images-v1/train.txt',
     ]
-    input_size = [224, 224]
+    class_name = None
+    input_size = [112, 112]
     rgb_mean = [0., 0., 0.]
     rgb_std = [1.0, 1.0, 1.0]
     transform = transforms.Compose([
@@ -254,9 +258,10 @@ if __name__ == '__main__':
     ])
     dataset = TextDataset(filename=filenames,
                           transform=transform,
-                          resample=True,
-                          shuffle=True,
-                          class_name=None,
+                          class_name=class_name,
+                          resample=False,
+                          shuffle=False,
+                          # crop_scale=(1.2, 1.2),
                           disp=True)
     for i in range(len(dataset)):
         data_info = dataset.__getitem__(i)

{pybaseutils-2.0.8 → pybaseutils-2.0.9}/pybaseutils.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pybaseutils
-Version: 2.0.8
+Version: 2.0.9
 Summary: pybaseutils
 Home-page: https://github.com/PanJinquan/base-utils
 Author: PanJinquan

{pybaseutils-2.0.8 → pybaseutils-2.0.9}/test_py/demo1.py

@@ -48,8 +48,10 @@ class Promote():
 
 
 if __name__ == '__main__':
-    file = "/home/PKing/nasdata/tmp/tmp/Natural/图片描述.json"
-    data = json_utils.read_json_data(file)
-    for i in range(len(data)):
-        data[i]["result"] = json.loads(data[i]["result"])
-    print(data)
+    image_file = "/home/PKing/nasdata/dataset-dmai/AIJE/dataset/aije-action-cvlm-v2/第3批数据/aije-action-train-v22/江门_检查工器具_20240910_172411_6606_0000.jpg"
+    boxe1 = [[600, 300, 850, 800]]
+    image = image_utils.read_image(image_file)
+    boxe2 = image_utils.get_square_boxes(boxe1, use_max=True, use_mean=True)
+    image = image_utils.draw_image_boxes(image, boxe1, color=(255, 0, 0))
+    image = image_utils.draw_image_boxes(image, boxe2, color=(0, 255, 0))
+    image_utils.cv_show_image("image", image)

{pybaseutils-2.0.8 → pybaseutils-2.0.9}/test_py/demo_rename.py

@@ -24,9 +24,9 @@ def rename_image_file(image_dir, output, prefix=""):
 
 if __name__ == '__main__':
     # image_dir = "/home/PKing/nasdata/dataset-dmai/AIJE/dataset/aije-indoor-det/dataset-v12/黄埔数企"
-    image_dir = "/home/PKing/Pictures/暗物16楼"
+    image_dir = "/home/PKing/nasdata/dataset-dmai/AIJE/dataset/aije-action-cvlm-v2/第3批数据/aije-action-train-v21"
     output = image_dir + "-train"
-    prefix = os.path.basename(image_dir)
+    prefix = "江门_更换熔丝"
     # image_dir = "/home/PKing/nasdata/dataset-dmai/AIJE/dataset/aije-indoor-cls2/手与表箱门有接触/手与表箱门无接触/南沙工匠"
     # output = "/home/PKing/nasdata/dataset-dmai/AIJE/岗评项目数据/室内考题/南沙视频/室内/南沙基地视频3/南沙工匠"
     # prefix = os.path.basename(output)