Python Tensorflow 2.0未安装在带有Heroku的pipenv上
我使用tf为django提供了pipenv锁,但它不会继续安装在heroku服务器上Python Tensorflow 2.0未安装在带有Heroku的pipenv上,python,tensorflow,Python,Tensorflow,我使用tf为django提供了pipenv锁,但它不会继续安装在heroku服务器上 An error occurred while installing tensorflow==2.0.0! Will try again. 我希望服务器可以安装2.0 tensorflow,但它失败了,而且没有事先的帮助 dj-database-url = "==0.5.0" django-storages = "==1.7.1" django = "==2.1.5" gunicorn = "==19.9.0
An error occurred while installing tensorflow==2.0.0! Will try again.
我希望服务器可以安装2.0 tensorflow,但它失败了,而且没有事先的帮助
dj-database-url = "==0.5.0"
django-storages = "==1.7.1"
django = "==2.1.5"
gunicorn = "==19.9.0"
psycopg2-binary = "==2.7.6.1"
psycopg2 = "==2.7.6.1"
markdownify = "==0.4.1"
numpy = "==1.16.2"
pipenv = "*"
regex = "==2017.4.5"
tensorboard = "*"
tb-nightly = "==1.14.0a20190301"
tf-estimator-nightly = "==1.14.0.dev2019030115"
tensorflow = "*"
tensorflow-estimator = "*"
Keras-Applications = "*"
Keras-Preprocessing = "*"
本地计算机将tensorflow安装到2.0.0
但是heroku机器需要2.0.0a0或b0或b1,它们不会在本地安装
pipenv安装tensorflow-v
显示:
requires absl-py>=0.7.0, astor>=0.6.0, gast==0.2.2, google-pasta>=0.1.6, grpcio>=1.8.6, keras-applications>=1.0.8, keras-preprocessing>=1.0.5, numpy<2.0,>=1.16.0, opt-einsum>=2.3.2, protobuf>=3.6.1, six>=1.10.0, tensorboard<2.1.0,>=2.0.0, tensorflow-estimator<2.1.0,>=2.0.0, termcolor>=1.1.0, wheel>=0.26, wrapt>=1.11.1
所需的软件包包括:
absl-py = "==0.7.1"
astor = "==0.7.1"
gast = "==0.2.2"
protobuf = "*"
six = "==1.12.0"
termcolor = "==1.1.0"
Tensorflow仅支持从python 3.4到python 3.7(截至2020年3月)。您可以在文档的[system requirements page][1]中找到更多信息 确保在PIP文件中包含以下内容:
[requires]
python_version = "3.7"
如果不是这样,请删除虚拟环境,将python\u版本更改为正确的版本,然后重新创建环境
我希望有帮助 如果您还没有看到任何错误,请在安装时尝试增加pipenv的详细程度(-v
),这是pipenv lock-v
还是pipenv install tensorflow-v
?pipenv--help
提供了用法:pipenv[OPTIONS]命令[ARGS]…
,因此我会尝试pipenv-v安装,这(如果未提供任何软件包),则从Pipfile安装所有软件包。
两个选项都不起作用
[requires]
python_version = "3.7"