Fix typo

hub_module/modules/image/classification/efficientnetb0_small_imagenet/module.py

@@ -24,9 +24,9 @@ from efficientnetb0_small_imagenet.efficientnet import EfficientNetB0_small
     author="paddlepaddle",
     author_email="paddle-dev@baidu.com",
     summary=
-    "ResNet18vd is a image classfication model, this module is trained with imagenet datasets.",
+    "EfficientNetB0 is a image classfication model, this module is trained with imagenet datasets.",
     version="1.0.0")
-class ResNet18vdImageNet(hub.Module):
+class EfficientNetB0ImageNet(hub.Module):
     def _initialize(self):
         self.default_pretrained_model_path = os.path.join(
             self.directory, "efficientnetb0_small_imagenet_model")
@@ -91,8 +91,8 @@ class ResNet18vdImageNet(hub.Module):
             with fluid.unique_name.guard():
                 image = fluid.layers.data(
                     name="image", shape=[3, 224, 224], dtype="float32")
-                resnet_vd = EfficientNetB0_small()
-                output, feature_map = resnet_vd.net(
+                efficientnet_b0 = EfficientNetB0_small()
+                output, feature_map = efficientnet_b0.net(
                     input=image, class_dim=len(self.label_list))
 
                 name_prefix = '@HUB_{}@'.format(self.name)
@@ -169,6 +169,15 @@ class ResNet18vdImageNet(hub.Module):
             self._set_config()
             self.predictor_set = True
 
+        if use_gpu:
+            try:
+                _places = os.environ["CUDA_VISIBLE_DEVICES"]
+                int(_places[0])
+            except:
+                raise RuntimeError(
+                    "Attempt to use GPU for prediction, but environment variable CUDA_VISIBLE_DEVICES was not set correctly."
+                )
+
         all_data = list()
         for yield_data in reader(images, paths):
             all_data.append(yield_data)

hub_module/modules/image/classification/fix_resnext101_32x48d_wsl_imagenet/module.py

@@ -155,6 +155,15 @@ class FixResnext10132x48dwslImagenet(hub.Module):
         Returns:
             res (list[dict]): The classfication results.
         """
+        if use_gpu:
+            try:
+                _places = os.environ["CUDA_VISIBLE_DEVICES"]
+                int(_places[0])
+            except:
+                raise RuntimeError(
+                    "Attempt to use GPU for prediction, but environment variable CUDA_VISIBLE_DEVICES was not set correctly."
+                )
+
         if not self.predictor_set:
             self._set_config()
             self.predictor_set = True

hub_module/modules/image/classification/mobilenet_v2_animals/module.py

@@ -155,6 +155,15 @@ class MobileNetV2Animals(hub.Module):
         Returns:
             res (list[dict]): The classfication results.
         """
+        if use_gpu:
+            try:
+                _places = os.environ["CUDA_VISIBLE_DEVICES"]
+                int(_places[0])
+            except:
+                raise RuntimeError(
+                    "Attempt to use GPU for prediction, but environment variable CUDA_VISIBLE_DEVICES was not set correctly."
+                )
+
         all_data = list()
         for yield_data in reader(images, paths):
             all_data.append(yield_data)

hub_module/modules/image/classification/mobilenet_v2_dishes/module.py

@@ -155,6 +155,15 @@ class MobileNetV2Dishes(hub.Module):
         Returns:
             res (list[dict]): The classfication results.
         """
+        if use_gpu:
+            try:
+                _places = os.environ["CUDA_VISIBLE_DEVICES"]
+                int(_places[0])
+            except:
+                raise RuntimeError(
+                    "Attempt to use GPU for prediction, but environment variable CUDA_VISIBLE_DEVICES was not set correctly."
+                )
+
         all_data = list()
         for yield_data in reader(images, paths):
             all_data.append(yield_data)

hub_module/modules/image/classification/mobilenet_v2_imagenet_ssld/module.py

@@ -178,6 +178,15 @@ class MobileNetV2ImageNetSSLD(hub.Module):
         Returns:
             res (list[dict]): The classfication results.
         """
+        if use_gpu:
+            try:
+                _places = os.environ["CUDA_VISIBLE_DEVICES"]
+                int(_places[0])
+            except:
+                raise RuntimeError(
+                    "Attempt to use GPU for prediction, but environment variable CUDA_VISIBLE_DEVICES was not set correctly."
+                )
+
         all_data = list()
         for yield_data in reader(images, paths):
             all_data.append(yield_data)

hub_module/modules/image/classification/mobilenet_v3_large_imagenet_ssld/module.py

@@ -155,6 +155,15 @@ class MobileNetV3Large(hub.Module):
         Returns:
             res (list[dict]): The classfication results.
         """
+        if use_gpu:
+            try:
+                _places = os.environ["CUDA_VISIBLE_DEVICES"]
+                int(_places[0])
+            except:
+                raise RuntimeError(
+                    "Attempt to use GPU for prediction, but environment variable CUDA_VISIBLE_DEVICES was not set correctly."
+                )
+
         all_data = list()
         for yield_data in reader(images, paths):
             all_data.append(yield_data)

hub_module/modules/image/classification/mobilenet_v3_small_imagenet_ssld/module.py

@@ -155,6 +155,15 @@ class MobileNetV3Small(hub.Module):
         Returns:
             res (list[dict]): The classfication results.
         """
+        if use_gpu:
+            try:
+                _places = os.environ["CUDA_VISIBLE_DEVICES"]
+                int(_places[0])
+            except:
+                raise RuntimeError(
+                    "Attempt to use GPU for prediction, but environment variable CUDA_VISIBLE_DEVICES was not set correctly."
+                )
+
         all_data = list()
         for yield_data in reader(images, paths):
             all_data.append(yield_data)

hub_module/modules/image/classification/resnet18_vd_imagenet/module.py

@@ -155,6 +155,15 @@ class ResNet18vdImageNet(hub.Module):
         Returns:
             res (list[dict]): The classfication results.
         """
+        if use_gpu:
+            try:
+                _places = os.environ["CUDA_VISIBLE_DEVICES"]
+                int(_places[0])
+            except:
+                raise RuntimeError(
+                    "Attempt to use GPU for prediction, but environment variable CUDA_VISIBLE_DEVICES was not set correctly."
+                )
+
         if not self.predictor_set:
             self._set_config()
             self.predictor_set = True

hub_module/modules/image/classification/resnet50_vd_animals/module.py

@@ -155,6 +155,15 @@ class ResNet50vdAnimals(hub.Module):
         Returns:
             res (list[dict]): The classfication results.
         """
+        if use_gpu:
+            try:
+                _places = os.environ["CUDA_VISIBLE_DEVICES"]
+                int(_places[0])
+            except:
+                raise RuntimeError(
+                    "Attempt to use GPU for prediction, but environment variable CUDA_VISIBLE_DEVICES was not set correctly."
+                )
+
         all_data = list()
         for yield_data in reader(images, paths):
             all_data.append(yield_data)

hub_module/modules/image/classification/resnet50_vd_dishes/module.py

@@ -155,6 +155,15 @@ class ResNet50vdDishes(hub.Module):
         Returns:
             res (list[dict]): The classfication results.
         """
+        if use_gpu:
+            try:
+                _places = os.environ["CUDA_VISIBLE_DEVICES"]
+                int(_places[0])
+            except:
+                raise RuntimeError(
+                    "Attempt to use GPU for prediction, but environment variable CUDA_VISIBLE_DEVICES was not set correctly."
+                )
+
         all_data = list()
         for yield_data in reader(images, paths):
             all_data.append(yield_data)

hub_module/modules/image/classification/resnet50_vd_imagenet_ssld/module.py

@@ -155,6 +155,15 @@ class ResNet50vdDishes(hub.Module):
         Returns:
             res (list[dict]): The classfication results.
         """
+        if use_gpu:
+            try:
+                _places = os.environ["CUDA_VISIBLE_DEVICES"]
+                int(_places[0])
+            except:
+                raise RuntimeError(
+                    "Attempt to use GPU for prediction, but environment variable CUDA_VISIBLE_DEVICES was not set correctly."
+                )
+
         all_data = list()
         for yield_data in reader(images, paths):
             all_data.append(yield_data)

hub_module/modules/image/classification/resnet50_vd_wildanimals/module.py

@@ -155,6 +155,15 @@ class ResNet50vdWildAnimals(hub.Module):
         Returns:
             res (list[dict]): The classfication results.
         """
+        if use_gpu:
+            try:
+                _places = os.environ["CUDA_VISIBLE_DEVICES"]
+                int(_places[0])
+            except:
+                raise RuntimeError(
+                    "Attempt to use GPU for prediction, but environment variable CUDA_VISIBLE_DEVICES was not set correctly."
+                )
+
         all_data = list()
         for yield_data in reader(images, paths):
             all_data.append(yield_data)

hub_module/modules/image/classification/se_resnet18_vd_imagenet/module.py

@@ -155,6 +155,15 @@ class SEResNet18vdImageNet(hub.Module):
         Returns:
             res (list[dict]): The classfication results.
         """
+        if use_gpu:
+            try:
+                _places = os.environ["CUDA_VISIBLE_DEVICES"]
+                int(_places[0])
+            except:
+                raise RuntimeError(
+                    "Attempt to use GPU for prediction, but environment variable CUDA_VISIBLE_DEVICES was not set correctly."
+                )
+
         if not self.predictor_set:
             self._set_config()
             self.predictor_set = True

hub_module/modules/image/object_detection/faster_rcnn_resnet50_coco2017/module.py

@@ -74,9 +74,9 @@ class FasterRCNNResNet50(hub.Module):
         Distill the Head Features, so as to perform transfer learning.
 
         Args:
+            num_classes (int): number of categories
             trainable (bool): whether to set parameters trainable.
             pretrained (bool): whether to load default pretrained model.
-            get_prediction (bool): whether to get prediction.
             phase (str): optional choices are 'train' and 'predict'.
 
         Returns:
@@ -263,6 +263,29 @@ class FasterRCNNResNet50(hub.Module):
             fg_thresh=0.5,
             class_nums=num_classes)
 
+    def save_inference_model(self,
+                             dirname,
+                             model_filename=None,
+                             params_filename=None,
+                             combined=True):
+        if combined:
+            model_filename = "__model__" if not model_filename else model_filename
+            params_filename = "__params__" if not params_filename else params_filename
+        place = fluid.CPUPlace()
+        exe = fluid.Executor(place)
+
+        program, feeded_var_names, target_vars = fluid.io.load_inference_model(
+            dirname=self.default_pretrained_model_path, executor=exe)
+
+        fluid.io.save_inference_model(
+            dirname=dirname,
+            main_program=program,
+            executor=exe,
+            feeded_var_names=feeded_var_names,
+            target_vars=target_vars,
+            model_filename=model_filename,
+            params_filename=params_filename)
+
     def object_detection(self,
                          paths=None,
                          images=None,
@@ -294,6 +317,14 @@ class FasterRCNNResNet50(hub.Module):
                     confidence (float): The confidence of detection result.
                 save_path (str, optional): The path to save output images.
         """
+        if use_gpu:
+            try:
+                _places = os.environ["CUDA_VISIBLE_DEVICES"]
+                int(_places[0])
+            except:
+                raise RuntimeError(
+                    "Attempt to use GPU for prediction, but environment variable CUDA_VISIBLE_DEVICES was not set correctly."
+                )
         paths = paths if paths else list()
         if data and 'image' in data:
             paths += data['image']
@@ -384,7 +415,7 @@ class FasterRCNNResNet50(hub.Module):
         Run as a service.
         """
         images_decode = [base64_to_cv2(image) for image in images]
-        results = self.object_detection(images_decode, **kwargs)
+        results = self.object_detection(images=images_decode, **kwargs)
         return results
 
     @runnable

hub_module/modules/image/object_detection/faster_rcnn_resnet50_coco2017/processor.py

@@ -80,7 +80,7 @@ def clip_bbox(bbox, img_width, img_height):
     ymin = max(min(bbox[1], img_height), 0.)
     xmax = max(min(bbox[2], img_width), 0.)
     ymax = max(min(bbox[3], img_height), 0.)
-    return xmin, ymin, xmax, ymax
+    return float(xmin), float(ymin), float(xmax), float(ymax)
 
 
 def load_label_info(file_path):
@@ -163,7 +163,7 @@ def postprocess(paths,
             bbox = row[2:]
             dt = {}
             dt['label'] = label_names[category_id]
-            dt['confidence'] = confidence
+            dt['confidence'] = float(confidence)
             dt['left'], dt['top'], dt['right'], dt['bottom'] = clip_bbox(
                 bbox, org_img_width, org_img_height)
             output_i['data'].append(dt)

hub_module/modules/image/object_detection/faster_rcnn_resnet50_fpn_coco2017/module.py

@@ -43,7 +43,7 @@ class FasterRCNNResNet50RPN(hub.Module):
             self.directory, "faster_rcnn_resnet50_fpn_model")
         self.label_names = load_label_info(
             os.path.join(self.directory, "label_file.txt"))
-        # self._set_config()
+        self._set_config()
 
     def _set_config(self):
         """
@@ -75,9 +75,9 @@ class FasterRCNNResNet50RPN(hub.Module):
         Distill the Head Features, so as to perform transfer learning.
 
         Args:
+            num_classes (int): number of categories
             trainable (bool): whether to set parameters trainable.
             pretrained (bool): whether to load default pretrained model.
-            get_prediction (bool): whether to get prediction.
             phase (str): optional choices are 'train' and 'predict'.
 
         Returns:
@@ -273,6 +273,29 @@ class FasterRCNNResNet50RPN(hub.Module):
             fg_thresh=0.5,
             class_nums=num_classes)
 
+    def save_inference_model(self,
+                             dirname,
+                             model_filename=None,
+                             params_filename=None,
+                             combined=True):
+        if combined:
+            model_filename = "__model__" if not model_filename else model_filename
+            params_filename = "__params__" if not params_filename else params_filename
+        place = fluid.CPUPlace()
+        exe = fluid.Executor(place)
+
+        program, feeded_var_names, target_vars = fluid.io.load_inference_model(
+            dirname=self.default_pretrained_model_path, executor=exe)
+
+        fluid.io.save_inference_model(
+            dirname=dirname,
+            main_program=program,
+            executor=exe,
+            feeded_var_names=feeded_var_names,
+            target_vars=target_vars,
+            model_filename=model_filename,
+            params_filename=params_filename)
+
     def object_detection(self,
                          paths=None,
                          images=None,
@@ -304,6 +327,15 @@ class FasterRCNNResNet50RPN(hub.Module):
                     confidence (float): The confidence of detection result.
                 save_path (str, optional): The path to save output images.
         """
+        if use_gpu:
+            try:
+                _places = os.environ["CUDA_VISIBLE_DEVICES"]
+                int(_places[0])
+            except:
+                raise RuntimeError(
+                    "Attempt to use GPU for prediction, but environment variable CUDA_VISIBLE_DEVICES was not set correctly."
+                )
+
         paths = paths if paths else list()
         if data and 'image' in data:
             paths += data['image']
@@ -399,7 +431,7 @@ class FasterRCNNResNet50RPN(hub.Module):
         Run as a service.
         """
         images_decode = [base64_to_cv2(image) for image in images]
-        results = self.object_detection(images_decode, **kwargs)
+        results = self.object_detection(images=images_decode, **kwargs)
         return results
 
     @runnable

hub_module/modules/image/object_detection/faster_rcnn_resnet50_fpn_coco2017/processor.py

@@ -80,7 +80,7 @@ def clip_bbox(bbox, img_width, img_height):
     ymin = max(min(bbox[1], img_height), 0.)
     xmax = max(min(bbox[2], img_width), 0.)
     ymax = max(min(bbox[3], img_height), 0.)
-    return xmin, ymin, xmax, ymax
+    return float(xmin), float(ymin), float(xmax), float(ymax)
 
 
 def load_label_info(file_path):
@@ -163,7 +163,7 @@ def postprocess(paths,
             bbox = row[2:]
             dt = {}
             dt['label'] = label_names[category_id]
-            dt['confidence'] = confidence
+            dt['confidence'] = float(confidence)
             dt['left'], dt['top'], dt['right'], dt['bottom'] = clip_bbox(
                 bbox, org_img_width, org_img_height)
             output_i['data'].append(dt)

hub_module/modules/image/object_detection/retinanet_resnet50_fpn_coco2017/module.py

@@ -14,6 +14,7 @@ import paddlehub as hub
 from paddlehub.module.module import moduleinfo, runnable, serving
 from paddle.fluid.core import PaddleTensor, AnalysisConfig, create_paddle_predictor
 from paddlehub.io.parser import txt_parser
+from paddlehub.common.paddle_helper import add_vars_prefix
 
 from retinanet_resnet50_fpn_coco2017.fpn import FPN
 from retinanet_resnet50_fpn_coco2017.retina_head import AnchorGenerator, RetinaTargetAssign, RetinaOutputDecoder, RetinaHead
@@ -86,10 +87,11 @@ class RetinaNetResNet50FPN(hub.Module):
         context_prog = fluid.Program()
         startup_program = fluid.Program()
         with fluid.program_guard(context_prog, startup_program):
+            var_prefix = '@HUB_{}@'.format(self.name)
             # image
             image = fluid.layers.data(
                 name='image',
-                shape=[3, 800, 1333],
+                shape=[-1, 3, -1, -1],
                 dtype='float32',
                 lod_level=0)
             # im_info
@@ -133,13 +135,31 @@ class RetinaNetResNet50FPN(hub.Module):
             # body_feats
             body_feats, spatial_scale = fpn.get_output(body_feats)
             # inputs, outputs, context_prog
-            inputs = {'image': image, 'im_info': im_info}
+            inputs = {
+                'image': var_prefix + image.name,
+                'im_info': var_prefix + im_info.name
+            }
             if get_prediction:
                 pred = retina_head.get_prediction(body_feats, spatial_scale,
                                                   im_info)
-                outputs = {'bbox_out': pred}
+                outputs = {'bbox_out': var_prefix + pred.name}
             else:
-                outputs = {'body_feats': body_feats}
+                outputs = {
+                    'body_features':
+                    [var_prefix + var.name for key, var in body_feats.items()]
+                }
+
+            # add_vars_prefix
+            add_vars_prefix(context_prog, var_prefix)
+            add_vars_prefix(fluid.default_startup_program(), var_prefix)
+
+            global_vars = context_prog.global_block().vars
+            inputs = {key: global_vars[value] for key, value in inputs.items()}
+            outputs = {
+                key: global_vars[value] if not isinstance(value, list) else
+                [global_vars[var] for var in value]
+                for key, value in outputs.items()
+            }
 
             place = fluid.CPUPlace()
             exe = fluid.Executor(place)
@@ -160,6 +180,29 @@ class RetinaNetResNet50FPN(hub.Module):
                 exe.run(startup_program)
             return inputs, outputs, context_prog
 
+    def save_inference_model(self,
+                             dirname,
+                             model_filename=None,
+                             params_filename=None,
+                             combined=True):
+        if combined:
+            model_filename = "__model__" if not model_filename else model_filename
+            params_filename = "__params__" if not params_filename else params_filename
+        place = fluid.CPUPlace()
+        exe = fluid.Executor(place)
+
+        program, feeded_var_names, target_vars = fluid.io.load_inference_model(
+            dirname=self.default_pretrained_model_path, executor=exe)
+
+        fluid.io.save_inference_model(
+            dirname=dirname,
+            main_program=program,
+            executor=exe,
+            feeded_var_names=feeded_var_names,
+            target_vars=target_vars,
+            model_filename=model_filename,
+            params_filename=params_filename)
+
     def object_detection(self,
                          paths=None,
                          images=None,
@@ -191,6 +234,15 @@ class RetinaNetResNet50FPN(hub.Module):
                     confidence (float): The confidence of detection result.
                 save_path (str, optional): The path to save output images.
         """
+        if use_gpu:
+            try:
+                _places = os.environ["CUDA_VISIBLE_DEVICES"]
+                int(_places[0])
+            except:
+                raise RuntimeError(
+                    "Attempt to use GPU for prediction, but environment variable CUDA_VISIBLE_DEVICES was not set correctly."
+                )
+
         all_images = list()
         paths = paths if paths else list()
         for yield_data in test_reader(paths, images):
@@ -274,7 +326,7 @@ class RetinaNetResNet50FPN(hub.Module):
         Run as a service.
         """
         images_decode = [base64_to_cv2(image) for image in images]
-        results = self.object_detection(images_decode, **kwargs)
+        results = self.object_detection(images=images_decode, **kwargs)
         return results
 
     @runnable

hub_module/modules/image/object_detection/retinanet_resnet50_fpn_coco2017/processor.py

@@ -76,7 +76,7 @@ def clip_bbox(bbox, img_width, img_height):
     ymin = max(min(bbox[1], img_height), 0.)
     xmax = max(min(bbox[2], img_width), 0.)
     ymax = max(min(bbox[3], img_height), 0.)
-    return xmin, ymin, xmax, ymax
+    return float(xmin), float(ymin), float(xmax), float(ymax)
 
 
 def load_label_info(file_path):
@@ -160,7 +160,7 @@ def postprocess(paths, images, data_out, score_thresh, label_names, output_dir,
             bbox = row[2:]
             dt = {}
             dt['label'] = label_names[category_id]
-            dt['confidence'] = confidence
+            dt['confidence'] = float(confidence)
             dt['left'], dt['top'], dt['right'], dt['bottom'] = clip_bbox(
                 bbox, org_img_width, org_img_height)
             output_i['data'].append(dt)

hub_module/modules/image/object_detection/ssd_mobilenet_v1_pascal/module.py

@@ -188,6 +188,15 @@ class SSDMobileNetv1(hub.Module):
                     confidence (float): The confidence of detection result.
                 save_path (str, optional): The path to save output images.
         """
+        if use_gpu:
+            try:
+                _places = os.environ["CUDA_VISIBLE_DEVICES"]
+                int(_places[0])
+            except:
+                raise RuntimeError(
+                    "Attempt to use GPU for prediction, but environment variable CUDA_VISIBLE_DEVICES was not set correctly."
+                )
+
         paths = paths if paths else list()
         if data and 'image' in data:
             paths += data['image']
@@ -244,7 +253,7 @@ class SSDMobileNetv1(hub.Module):
         Run as a service.
         """
         images_decode = [base64_to_cv2(image) for image in images]
-        results = self.object_detection(images_decode, **kwargs)
+        results = self.object_detection(images=images_decode, **kwargs)
         return results
 
     @runnable

hub_module/modules/image/object_detection/ssd_mobilenet_v1_pascal/processor.py

@@ -77,7 +77,7 @@ def clip_bbox(bbox, img_width, img_height):
     ymin = max(min(bbox[1], img_height), 0.)
     xmax = max(min(bbox[2], img_width), 0.)
     ymax = max(min(bbox[3], img_height), 0.)
-    return xmin, ymin, xmax, ymax
+    return float(xmin), float(ymin), float(xmax), float(ymax)
 
 
 def load_label_info(file_path):
@@ -163,7 +163,7 @@ def postprocess(paths,
             bbox[3] = bbox[3] * org_img_height
             dt = {}
             dt['label'] = label_names[category_id]
-            dt['confidence'] = confidence
+            dt['confidence'] = float(confidence)
             dt['left'], dt['top'], dt['right'], dt['bottom'] = clip_bbox(
                 bbox, org_img_width, org_img_height)
             output_i['data'].append(dt)

hub_module/modules/image/object_detection/ssd_vgg16_300_coco2017/module.py

@@ -242,7 +242,7 @@ class SSDVGG16(hub.Module):
         Run as a service.
         """
         images_decode = [base64_to_cv2(image) for image in images]
-        results = self.object_detection(images_decode, **kwargs)
+        results = self.object_detection(images=images_decode, **kwargs)
         return results
 
     @runnable

hub_module/modules/image/object_detection/ssd_vgg16_300_coco2017/processor.py

@@ -77,7 +77,7 @@ def clip_bbox(bbox, img_width, img_height):
     ymin = max(min(bbox[1], img_height), 0.)
     xmax = max(min(bbox[2], img_width), 0.)
     ymax = max(min(bbox[3], img_height), 0.)
-    return xmin, ymin, xmax, ymax
+    return float(xmin), float(ymin), float(xmax), float(ymax)
 
 
 def load_label_info(file_path):
@@ -163,7 +163,7 @@ def postprocess(paths,
             bbox[3] = bbox[3] * org_img_height
             dt = {}
             dt['label'] = label_names[category_id]
-            dt['confidence'] = confidence
+            dt['confidence'] = float(confidence)
             dt['left'], dt['top'], dt['right'], dt['bottom'] = clip_bbox(
                 bbox, org_img_width, org_img_height)
             output_i['data'].append(dt)

hub_module/modules/image/object_detection/ssd_vgg16_512_coco2017/module.py

@@ -194,6 +194,15 @@ class SSDVGG16_512(hub.Module):
                     confidence (float): The confidence of detection result.
                 save_path (str, optional): The path to save output images.
         """
+        if use_gpu:
+            try:
+                _places = os.environ["CUDA_VISIBLE_DEVICES"]
+                int(_places[0])
+            except:
+                raise RuntimeError(
+                    "Attempt to use GPU for prediction, but environment variable CUDA_VISIBLE_DEVICES was not set correctly."
+                )
+
         paths = paths if paths else list()
         data_reader = partial(reader, paths, images)
         batch_reader = fluid.io.batch(data_reader, batch_size=batch_size)
@@ -247,7 +256,7 @@ class SSDVGG16_512(hub.Module):
         Run as a service.
         """
         images_decode = [base64_to_cv2(image) for image in images]
-        results = self.object_detection(images_decode, **kwargs)
+        results = self.object_detection(images=images_decode, **kwargs)
         return results
 
     @runnable

hub_module/modules/image/object_detection/ssd_vgg16_512_coco2017/processor.py

@@ -77,7 +77,7 @@ def clip_bbox(bbox, img_width, img_height):
     ymin = max(min(bbox[1], img_height), 0.)
     xmax = max(min(bbox[2], img_width), 0.)
     ymax = max(min(bbox[3], img_height), 0.)
-    return xmin, ymin, xmax, ymax
+    return float(xmin), float(ymin), float(xmax), float(ymax)
 
 
 def load_label_info(file_path):
@@ -163,7 +163,7 @@ def postprocess(paths,
             bbox[3] = bbox[3] * org_img_height
             dt = {}
             dt['label'] = label_names[category_id]
-            dt['confidence'] = confidence
+            dt['confidence'] = float(confidence)
             dt['left'], dt['top'], dt['right'], dt['bottom'] = clip_bbox(
                 bbox, org_img_width, org_img_height)
             output_i['data'].append(dt)

hub_module/modules/image/object_detection/yolov3_darknet53_coco2017/module.py

@@ -180,6 +180,15 @@ class YOLOv3DarkNet53Coco2017(hub.Module):
                     confidence (float): The confidence of detection result.
                 save_path (str, optional): The path to save output images.
         """
+        if use_gpu:
+            try:
+                _places = os.environ["CUDA_VISIBLE_DEVICES"]
+                int(_places[0])
+            except:
+                raise RuntimeError(
+                    "Attempt to use GPU for prediction, but environment variable CUDA_VISIBLE_DEVICES was not set correctly."
+                )
+
         paths = paths if paths else list()
         if data and 'image' in data:
             paths += data['image']
@@ -239,7 +248,7 @@ class YOLOv3DarkNet53Coco2017(hub.Module):
         Run as a service.
         """
         images_decode = [base64_to_cv2(image) for image in images]
-        results = self.object_detection(images_decode, **kwargs)
+        results = self.object_detection(images=images_decode, **kwargs)
         return results
 
     @runnable

hub_module/modules/image/object_detection/yolov3_darknet53_coco2017/processor.py

@@ -80,7 +80,7 @@ def clip_bbox(bbox, img_width, img_height):
     ymin = max(min(bbox[1], img_height), 0.)
     xmax = max(min(bbox[2], img_width), 0.)
     ymax = max(min(bbox[3], img_height), 0.)
-    return xmin, ymin, xmax, ymax
+    return float(xmin), float(ymin), float(xmax), float(ymax)
 
 
 def load_label_info(file_path):
@@ -167,7 +167,7 @@ def postprocess(paths,
             bbox = row[2:]
             dt = {}
             dt['label'] = label_names[category_id]
-            dt['confidence'] = confidence
+            dt['confidence'] = float(confidence)
             dt['left'], dt['top'], dt['right'], dt['bottom'] = clip_bbox(
                 bbox, org_img_width, org_img_height)
             output_i['data'].append(dt)

hub_module/modules/image/object_detection/yolov3_darknet53_pedestrian/module.py

@@ -193,6 +193,15 @@ class YOLOv3DarkNet53Pedestrian(hub.Module):
                     confidence (float): The confidence of detection result.
                 save_path (str, optional): The path to save output images.
         """
+        if use_gpu:
+            try:
+                _places = os.environ["CUDA_VISIBLE_DEVICES"]
+                int(_places[0])
+            except:
+                raise RuntimeError(
+                    "Attempt to use GPU for prediction, but environment variable CUDA_VISIBLE_DEVICES was not set correctly."
+                )
+
         paths = paths if paths else list()
         data_reader = partial(reader, paths, images)
         batch_reader = fluid.io.batch(data_reader, batch_size=batch_size)
@@ -249,7 +258,7 @@ class YOLOv3DarkNet53Pedestrian(hub.Module):
         Run as a service.
         """
         images_decode = [base64_to_cv2(image) for image in images]
-        results = self.object_detection(images_decode, **kwargs)
+        results = self.object_detection(images=images_decode, **kwargs)
         return results
 
     @runnable

hub_module/modules/image/object_detection/yolov3_darknet53_pedestrian/processor.py

@@ -80,7 +80,7 @@ def clip_bbox(bbox, img_width, img_height):
     ymin = max(min(bbox[1], img_height), 0.)
     xmax = max(min(bbox[2], img_width), 0.)
     ymax = max(min(bbox[3], img_height), 0.)
-    return xmin, ymin, xmax, ymax
+    return float(xmin), float(ymin), float(xmax), float(ymax)
 
 
 def load_label_info(file_path):
@@ -167,7 +167,7 @@ def postprocess(paths,
             bbox = row[2:]
             dt = {}
             dt['label'] = label_names[category_id]
-            dt['confidence'] = confidence
+            dt['confidence'] = float(confidence)
             dt['left'], dt['top'], dt['right'], dt['bottom'] = clip_bbox(
                 bbox, org_img_width, org_img_height)
             output_i['data'].append(dt)

hub_module/modules/image/object_detection/yolov3_darknet53_vehicles/module.py

@@ -193,6 +193,15 @@ class YOLOv3DarkNet53Vehicles(hub.Module):
                     confidence (float): The confidence of detection result.
                 save_path (str, optional): The path to save output images.
         """
+        if use_gpu:
+            try:
+                _places = os.environ["CUDA_VISIBLE_DEVICES"]
+                int(_places[0])
+            except:
+                raise RuntimeError(
+                    "Attempt to use GPU for prediction, but environment variable CUDA_VISIBLE_DEVICES was not set correctly."
+                )
+
         paths = paths if paths else list()
         data_reader = partial(reader, paths, images)
         batch_reader = fluid.io.batch(data_reader, batch_size=batch_size)
@@ -249,7 +258,7 @@ class YOLOv3DarkNet53Vehicles(hub.Module):
         Run as a service.
         """
         images_decode = [base64_to_cv2(image) for image in images]
-        results = self.object_detection(images_decode, **kwargs)
+        results = self.object_detection(images=images_decode, **kwargs)
         return results
 
     @runnable

hub_module/modules/image/object_detection/yolov3_darknet53_vehicles/processor.py

@@ -80,7 +80,7 @@ def clip_bbox(bbox, img_width, img_height):
     ymin = max(min(bbox[1], img_height), 0.)
     xmax = max(min(bbox[2], img_width), 0.)
     ymax = max(min(bbox[3], img_height), 0.)
-    return xmin, ymin, xmax, ymax
+    return float(xmin), float(ymin), float(xmax), float(ymax)
 
 
 def load_label_info(file_path):
@@ -167,7 +167,7 @@ def postprocess(paths,
             bbox = row[2:]
             dt = {}
             dt['label'] = label_names[category_id]
-            dt['confidence'] = confidence
+            dt['confidence'] = float(confidence)
             dt['left'], dt['top'], dt['right'], dt['bottom'] = clip_bbox(
                 bbox, org_img_width, org_img_height)
             output_i['data'].append(dt)

hub_module/modules/image/object_detection/yolov3_mobilenet_v1_coco2017/module.py

@@ -183,6 +183,15 @@ class YOLOv3MobileNetV1Coco2017(hub.Module):
                     confidence (float): The confidence of detection result.
                 save_path (str, optional): The path to save output images.
         """
+        if use_gpu:
+            try:
+                _places = os.environ["CUDA_VISIBLE_DEVICES"]
+                int(_places[0])
+            except:
+                raise RuntimeError(
+                    "Attempt to use GPU for prediction, but environment variable CUDA_VISIBLE_DEVICES was not set correctly."
+                )
+
         paths = paths if paths else list()
         data_reader = partial(reader, paths, images)
         batch_reader = fluid.io.batch(data_reader, batch_size=batch_size)
@@ -239,7 +248,7 @@ class YOLOv3MobileNetV1Coco2017(hub.Module):
         Run as a service.
         """
         images_decode = [base64_to_cv2(image) for image in images]
-        results = self.face_detection(images_decode, **kwargs)
+        results = self.object_detection(images=images_decode, **kwargs)
         return results
 
     @runnable

hub_module/modules/image/object_detection/yolov3_mobilenet_v1_coco2017/processor.py

@@ -80,7 +80,7 @@ def clip_bbox(bbox, img_width, img_height):
     ymin = max(min(bbox[1], img_height), 0.)
     xmax = max(min(bbox[2], img_width), 0.)
     ymax = max(min(bbox[3], img_height), 0.)
-    return xmin, ymin, xmax, ymax
+    return float(xmin), float(ymin), float(xmax), float(ymax)
 
 
 def load_label_info(file_path):
@@ -167,7 +167,7 @@ def postprocess(paths,
             bbox = row[2:]
             dt = {}
             dt['label'] = label_names[category_id]
-            dt['confidence'] = confidence
+            dt['confidence'] = float(confidence)
             dt['left'], dt['top'], dt['right'], dt['bottom'] = clip_bbox(
                 bbox, org_img_width, org_img_height)
             output_i['data'].append(dt)

hub_module/modules/image/object_detection/yolov3_resnet34_coco2017/module.py

@@ -185,6 +185,15 @@ class YOLOv3DarkNet53Coco2017(hub.Module):
                     confidence (float): The confidence of detection result.
                 save_path (str, optional): The path to save output images.
         """
+        if use_gpu:
+            try:
+                _places = os.environ["CUDA_VISIBLE_DEVICES"]
+                int(_places[0])
+            except:
+                raise RuntimeError(
+                    "Attempt to use GPU for prediction, but environment variable CUDA_VISIBLE_DEVICES was not set correctly."
+                )
+
         paths = paths if paths else list()
         data_reader = partial(reader, paths, images)
         batch_reader = fluid.io.batch(data_reader, batch_size=batch_size)
@@ -241,7 +250,7 @@ class YOLOv3DarkNet53Coco2017(hub.Module):
         Run as a service.
         """
         images_decode = [base64_to_cv2(image) for image in images]
-        results = self.object_detection(images_decode, **kwargs)
+        results = self.object_detection(images=images_decode, **kwargs)
         return results
 
     @runnable

hub_module/modules/image/object_detection/yolov3_resnet34_coco2017/processor.py

@@ -80,7 +80,7 @@ def clip_bbox(bbox, img_width, img_height):
     ymin = max(min(bbox[1], img_height), 0.)
     xmax = max(min(bbox[2], img_width), 0.)
     ymax = max(min(bbox[3], img_height), 0.)
-    return xmin, ymin, xmax, ymax
+    return float(xmin), float(ymin), float(xmax), float(ymax)
 
 
 def load_label_info(file_path):
@@ -167,7 +167,7 @@ def postprocess(paths,
             bbox = row[2:]
             dt = {}
             dt['label'] = label_names[category_id]
-            dt['confidence'] = confidence
+            dt['confidence'] = float(confidence)
             dt['left'], dt['top'], dt['right'], dt['bottom'] = clip_bbox(
                 bbox, org_img_width, org_img_height)
             output_i['data'].append(dt)