目标检测环境搭建
2020-12-12 本文已影响0人
h小吴y
github总地址:https://github.com/tensorflow/models
目标检测地址:https://github.com/tensorflow/models/tree/master/research/object_detection!

这里选择tensorflow2的版本安装,会跳转到下面目录:

第一步,先克隆整个库到本地(没有git,先安装):
git clone https://github.com/tensorflow/models.git
第二步,选择Python Package Installation安装:
# 进入research目录执行下面操作
cd models/research
# 编译protos(没有protobuf-compiler需要先安装).
protoc object_detection/protos/*.proto --python_out=.
# 安装需要的python库
cp object_detection/packages/tf2/setup.py .
python -m pip install --use-feature=2020-resolver .
最后,安装完成测试:
python object_detection/builders/model_builder_tf2_test.py