Merge pull request #225 from PaddlePaddle/develop

fix parameters

docs/examples/meter_reader.md

@@ -179,7 +179,7 @@ git clone https://github.com/PaddlePaddle/PaddleX
   .\paddlex_inference\meter_reader.exe --det_model_dir=\path\to\encrypted_det_inference_model --seg_model_dir=\path\to\encrypted_seg_inference_model --image=\path\to\test.jpg --use_gpu=1 --use_erode=1 --save_dir=output --det_key yEBLDiBOdlj+5EsNNrABhfDuQGkdcreYcHcncqwdbx0= --seg_key DbVS64I9pFRo5XmQ8MNV2kSGsfEr4FKA6OH9OUhRrsY=
   ```
 
-### Linux系统的jeton嵌入式设备安全部署
+### Linux系统的jetson嵌入式设备安全部署
 
 #### c++部署
 

docs/install.md

@@ -30,7 +30,8 @@ python setup.py install
 
 PaddleX依赖pycocotools包，如安装pycocotools失败，可参照如下方式安装pycocotools
 
-* Windows安装时可能会提示缺少`Microsoft Visual C++ 2015 build tools`，[点击下载VC build tools](https://go.microsoft.com/fwlink/?LinkId=691126)安装再执行如下pip命令
+* Windows安装时可能会提示`Microsoft Visual C++ 14.0 is required`，从而导致安装出错，[点击下载VC build tools](https://go.microsoft.com/fwlink/?LinkId=691126)安装再执行如下pip命令
+> 注意：安装完后，需要重新打开新的终端命令窗口
 
 ```
 pip install cython

docs/train/prediction.md

@@ -77,7 +77,8 @@ import paddlex as pdx
 test_jpg = './deeplabv3p_mobilenetv2_voc/test.jpg'
 model = pdx.load_model('./deeplabv3p_mobilenetv2_voc')
 result = model.predict(test_jpg)
-pdx.seg.visualize(test_jpg, result, weight=0.6, save_dir='./')
+# 可视化结果存储在./visualized_test.jpg，见下图右(左图为原图)
+pdx.seg.visualize(test_jpg, result, weight=0.0, save_dir='./')
 ```
 
 在上述示例代码中，通过调用`paddlex.seg.visualize`可以对语义分割的预测结果进行可视化，可视化的结果保存在`save_dir`下，见下图。其中`weight`参数用于调整预测结果和原图结果融合展现时的权重，0.0时只展示预测结果mask的可视化，1.0时只展示原图可视化。

examples/meter_reader/README.md

@@ -188,7 +188,7 @@ git clone https://github.com/PaddlePaddle/PaddleX
   .\paddlex_inference\meter_reader.exe --det_model_dir=\path\to\encrypted_det_inference_model --seg_model_dir=\path\to\encrypted_seg_inference_model --image=\path\to\test.jpg --use_gpu=1 --use_erode=1 --save_dir=output --det_key yEBLDiBOdlj+5EsNNrABhfDuQGkdcreYcHcncqwdbx0= --seg_key DbVS64I9pFRo5XmQ8MNV2kSGsfEr4FKA6OH9OUhRrsY=
   ```
 
-### Linux系统的jeton嵌入式设备安全部署
+### Linux系统的jetson嵌入式设备安全部署
 
 #### c++部署
 

examples/meter_reader/deploy/python/reader_deploy.py

@@ -335,7 +335,7 @@ def infer(args):
             meter_reader.predict(im_file, args.save_dir, args.use_erode,
                                  args.erode_kernel, args.score_threshold,
                                  args.seg_batch_size, args.seg_thread_num)
-    elif args.with_camera:
+    elif args.use_camera:
         cap_video = cv2.VideoCapture(args.camera_id)
         if not cap_video.isOpened():
             raise Exception(

examples/meter_reader/reader_infer.py

@@ -335,7 +335,7 @@ def infer(args):
             meter_reader.predict(im_file, args.save_dir, args.use_erode,
                                  args.erode_kernel, args.score_threshold,
                                  args.seg_batch_size, args.seg_thread_num)
-    elif args.with_camera:
+    elif args.use_camera:
         cap_video = cv2.VideoCapture(args.camera_id)
         if not cap_video.isOpened():
             raise Exception(

examples/meter_reader/train_detection.py

@@ -6,7 +6,7 @@ from paddlex.det import transforms
 import paddlex as pdx
 
 # 下载和解压表计检测数据集
-meter_det_dataset = 'https://bj.bcebos.com/paddlex/meterreader/datasets/meter_det.tar.gz'
+meter_det_dataset = 'https://bj.bcebos.com/paddlex/examples/meter_reader/datasets/meter_det.tar.gz'
 pdx.utils.download_and_decompress(meter_det_dataset, path='./')
 
 # 定义训练和验证时的transforms

examples/meter_reader/train_segmentation.py

@@ -6,7 +6,7 @@ import paddlex as pdx
 from paddlex.seg import transforms
 
 # 下载和解压表盘分割数据集
-meter_seg_dataset = 'https://bj.bcebos.com/paddlex/meterreader/datasets/meter_seg.tar.gz'
+meter_seg_dataset = 'https://bj.bcebos.com/paddlex/examples/meter_reader/datasets/meter_seg.tar.gz'
 pdx.utils.download_and_decompress(meter_seg_dataset, path='./')
 
 # 定义训练和验证时的transforms
@@ -42,8 +42,7 @@ eval_dataset = pdx.datasets.SegDataset(
 #
 # API说明: https://paddlex.readthedocs.io/zh_CN/latest/apis/models/semantic_segmentation.html#deeplabv3p
 model = pdx.seg.DeepLabv3p(
-    num_classes=len(train_dataset.labels),
-    backbone='Xception65')
+    num_classes=len(train_dataset.labels), backbone='Xception65')
 model.train(
     num_epochs=20,
     train_dataset=train_dataset,