英文:
a question about "ERROR: pip's dependency" for deep learning
问题
ERROR: pip的依赖解析器目前不考虑所有已安装的软件包。这个行为是以下依赖冲突的根源。
googleapis-common-protos 1.59.1需要protobuf!=3.20.0,!=3.20.1,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5，<5.0.0.dev0，但您安装了不兼容的protobuf 3.20.0。
tensorflow-metadata 1.13.1需要protobuf<5,>=3.20.3，但您安装了不兼容的protobuf 3.20.0。
tensorboard 2.10.1需要protobuf<3.20,>=3.9.2，但您安装了不兼容的protobuf 3.20.0。
tensorflow 2.10.1需要protobuf<3.20,>=3.9.2，但您安装了不兼容的protobuf 3.20.0。
我仍然可以运行我的深度学习代码。我需要解决上述问题吗？
谢谢
Erick
英文:
all,
After I installed a package with "pip install keras-flops", I had the following message.
ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
googleapis-common-protos 1.59.1 requires protobuf!=3.20.0,!=3.20.1,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0.dev0,>=3.19.5, but you have protobuf 3.20.0 which is incompatible.
tensorflow-metadata 1.13.1 requires protobuf<5,>=3.20.3, but you have protobuf 3.20.0 which is incompatible.
tensorboard 2.10.1 requires protobuf<3.20,>=3.9.2, but you have protobuf 3.20.0 which is incompatible.
tensorflow 2.10.1 requires protobuf<3.20,>=3.9.2, but you have protobuf 3.20.0 which is incompatible.
I can still run my deep learning codes. Do I need to fix the above problems?
Thanks
Erick
答案1
得分: 0
Keras-flops 是一个较旧的包，最后发布日期为：2020年8月17日，仅兼容于 tensorflow 和 keras 版本2.2，并需要使用 python 3.6。因此，在最新的tensorflow版本中安装此包会导致版本不兼容或pip依赖错误。请参考此 keras-flops pypi 链接获取更多详细信息。
您可以尝试使用最新的Tensorflow版本中的Tensorflow模型的 try_count_flops 来计算并返回模型的FLOPs。
英文:
Keras-flops is older package which was last released on : Aug 17, 2020 and only compatible with tensorflow and keras version 2.2 with python 3.6. So installing this package in latest tensorflow version is causing the version incompatibility or pip dependency error. Please refer to this keras-flops pypi link for more details.
You can alternatively try using Tensorflow Model's try_count_flops in the latest Tensorflow version which is used to count and return model FLOPs.
通过集体智慧和协作来改善编程学习和解决问题的方式。致力于成为全球开发者共同参与的知识库，让每个人都能够通过互相帮助和分享经验来进步。
评论