?在使用TensorFlow的object_detection模塊問(wèn)題記錄：

• 配置：window10
• tensorflow-gpu
• protoc3.8版本下載地址：https://github.com/protocolbuffers/protobuf/releases

1. ?ImportError: No module named 'object_detection' 問(wèn)題：

①將自己下載的models-master模塊添加進(jìn)去系統(tǒng)環(huán)境變量

②在F:\Anaconda\envs\自己環(huán)境下的\Lib\site-packages ?這個(gè)文件夾目錄下，寫一個(gè).pth文件，內(nèi)容為兩條路徑

2.protoc.exe運(yùn)行報(bào)錯(cuò)：

同樣，將protoc.exe添加進(jìn)系統(tǒng)環(huán)境變量

3."protoc object_detection/protos/*.proto --python_out=."命令錯(cuò)誤：

object_detection/protos/*.proto: No such file or directory

4.cannot import name 'anchor_generator_pb2' from 'object_detection.protos

問(wèn)題3和問(wèn)題4都是屬于下面對(duì)應(yīng)的一系列*pb2.py文件沒(méi)有生成成功

解決方法：

我的是protoc3.8版本，網(wǎng)上說(shuō)下載protoc3.4版本就不會(huì)報(bào)錯(cuò)，但是我的還是有問(wèn)題，最后我在github的issues找到了一個(gè)解決方法：

https://github.com/tensorflow/models/issues/2930

就是一個(gè)一個(gè)的CMD執(zhí)行（直接復(fù)制下面的30行代碼到CMD可以一次性全執(zhí)行完）把 *.proto文件執(zhí)行生成 ***pb2.py文件。

方便復(fù)制粘貼直接貼上：

            
              protoc object_detection/protos/anchor_generator.proto --python_out=.
protoc object_detection/protos/argmax_matcher.proto --python_out=.
protoc object_detection/protos/bipartite_matcher.proto --python_out=.
protoc object_detection/protos/box_coder.proto --python_out=.
protoc object_detection/protos/box_predictor.proto --python_out=.
protoc object_detection/protos/calibration.proto --python_out=.
protoc object_detection/protos/eval.proto --python_out=.
protoc object_detection/protos/faster_rcnn.proto --python_out=.
protoc object_detection/protos/faster_rcnn_box_coder.proto --python_out=.
protoc object_detection/protos/grid_anchor_generator.proto --python_out=.
protoc object_detection/protos/hyperparams.proto --python_out=.
protoc object_detection/protos/image_resizer.proto --python_out=.
protoc object_detection/protos/input_reader.proto --python_out=.
protoc object_detection/protos/keypoint_box_coder.proto --python_out=.
protoc object_detection/protos/losses.proto --python_out=.
protoc object_detection/protos/matcher.proto --python_out=.
protoc object_detection/protos/mean_stddev_box_coder.proto --python_out=.
protoc object_detection/protos/model.proto --python_out=.
protoc object_detection/protos/multiscale_anchor_generator.proto --python_out=.
protoc object_detection/protos/optimizer.proto --python_out=.
protoc object_detection/protos/pipeline.proto --python_out=.
protoc object_detection/protos/post_processing.proto --python_out=.
protoc object_detection/protos/preprocessor.proto --python_out=.
protoc object_detection/protos/region_similarity_calculator.proto --python_out=.
protoc object_detection/protos/square_box_coder.proto --python_out=.
protoc object_detection/protos/ssd.proto --python_out=.
protoc object_detection/protos/ssd_anchor_generator.proto --python_out=.
protoc object_detection/protos/string_int_label_map.proto --python_out=.
protoc object_detection/protos/train.proto --python_out=.
            
          

測(cè)試是否成功：

            
              python object_detection/builders/model_builder_test.py
            
          

參考文獻(xiàn)：

object_detection配置相關(guān):https://blog.csdn.net/Zlase/article/details/78734138