Comments (15)
I'm pretty sure the error comes from Anaconda. Can you try the following for me:
pip install --extra-index-url=https://pypi.ngc.nvidia.com --trusted-host pypi.ngc.nvidia.com nvidia-tensorflow
The objective of the package you try nvidia-pyindex
and that fails, install the nvidia python package repository to avoid having to write this complex command.
So if it doesn't work, you can go around it like that.
If you are using a docker container, can you please send me the docker container name ? I'd like to be able to reproduce it. Thanks
from tensorflow.
Thanks for your reply. Looks like I get the same error message:
$ pip install --extra-index-url=https://pypi.ngc.nvidia.com --trusted-host pypi.ngc.nvidia.com nvidia-tensorflow
Looking in indexes: https://pypi.org/simple, https://pypi.ngc.nvidia.com, https://pypi.ngc.nvidia.com
Collecting nvidia-tensorflow
Downloading nvidia-tensorflow-0.0.1.dev4.tar.gz (3.8 kB)
ERROR: Command errored out with exit status 1:
command: /home/shawley/anaconda3/envs/spnet/bin/python -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/tmp/pip-install-vd7fm7lc/nvidia-tensorflow/setup.py'"'"'; __file__='"'"'/tmp/pip-install-vd7fm7lc/nvidia-tensorflow/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(__file__);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' egg_info --egg-base /tmp/pip-pip-egg-info-ir6r10an
cwd: /tmp/pip-install-vd7fm7lc/nvidia-tensorflow/
Complete output (17 lines):
Traceback (most recent call last):
File "<string>", line 1, in <module>
File "/tmp/pip-install-vd7fm7lc/nvidia-tensorflow/setup.py", line 150, in <module>
raise RuntimeError(open("ERROR.txt", "r").read())
RuntimeError:
###########################################################################################
The package you are trying to install is only a placeholder project on PyPI.org repository.
This package is hosted on NVIDIA Python Package Index.
This package can be installed as:
```
$ pip install nvidia-pyindex
$ pip install nvidia-tensorflow
```
Please refer to NVIDIA instructions: https://github.com/NVIDIA/tensorflow#install.
###########################################################################################
----------------------------------------
ERROR: Command errored out with exit status 1: python setup.py egg_info Check the logs for full command output.
I'm able to install other packages via conda's pip, just not this package.
I tried deactivating the conda environment and starting a clean virtualenv environment, but am having other, more general, problems with pip supplied in the venv.
from tensorflow.
Hi,
thanks for allowing us to know that nvidia-tensorflow is only working with python3.8. Save my time a lot .
Following this instruction , I did not see that requirement in the webpage. I would suggest that maybe the webpage can indicate this in the pre-requirement section.
from tensorflow.
EDIT: I just realized you are using Python 3.7. We don't support that.
Only Python3.8 from the December release, and Python 3.6 for November and before
So everything is actually normal. PIP complains he can't find a candidate to install because None exists.
from tensorflow.
Ok problem now resolved. Had a couple issues along the way that I will post here for future users who may have similar issues:
With python 3.8 & venv/pip I was getting a different error.
$ python3.8 -m venv env_spnet
$ source env_spnet/bin/activate
(env_spnet) $ pip install --user nvidia-pyindex
Exception:
Traceback (most recent call last):
File "/home/shawley/env_spnet/lib/python3.8/site-packages/pip/basecommand.py", line 215, in main
status = self.run(options, args)
File "/home/shawley/env_spnet/lib/python3.8/site-packages/pip/commands/install.py", line 290, in run
with self._build_session(options) as session:
File "/home/shawley/env_spnet/lib/python3.8/site-packages/pip/basecommand.py", line 68, in _build_session
normalize_path(os.path.join(options.cache_dir, "http"))
File "/usr/lib/python3.8/posixpath.py", line 76, in join
a = os.fspath(a)
TypeError: expected str, bytes or os.PathLike object, not in
The problem here was that the pip installed by default on Ubuntu is ancient (v9.0). But if you bootstrap pip via
curl https://bootstrap.pypa.io/get-pip.py | python3.8 -
Then you can run the install commands, albeit without the --user
part.
After that, installation proceeds without errors. Closing this issue.
from tensorflow.
I have the same issue under Ubuntu 22.04.1 LTS with Python 3.10.6
Trying:
pip install nvidia-pyindex
pip install nvidia-tensorflow[horovod]
or
pip install --extra-index-url=https://pypi.ngc.nvidia.com --trusted-host pypi.ngc.nvidia.com nvidia-tensorflow
Both return the error previously shown which advises the package is a pypi.org placeholder and to see the install instructions.
EDIT: I notice that https://pypi.org/simple returns HTML that appears to be a list of packages, whereas https://pypi.ngc.nvidia.com redirects to https://developer.download.nvidia.com/compute/redist/ (which is a blank page)
I'm unsure whether the pypi.org HTML is there simply for the benefit of humans, or that's the actual index data that pip uses. If it's the latter, that may explain why pip cannot find nvidia-tensorflow on pypi.ngc.nvidia.com.
from tensorflow.
My apologies, I forgot to follow this up.
https://developer.download.nvidia.com/compute/redist/nvidia-tensorflow/ shows the list of WHL files.
Based on those filenames, it seems that "Only Python3.8" from an earlier reply is probably meant to be taken literally? :) ie, no support for 3.9 or later.
I set up a virtual environment with Python 3.8.15 and was able to install nvidia-tensorflow 1.15.5+nv22.11
from tensorflow.
Try to upgrade pip/setuptools/wheel: Ensure that you have the latest versions of pip, setuptools, and wheel installed. You can upgrade them using the following command:
pip install --upgrade pip setuptools wheel
from tensorflow.
$ pip --version
pip 20.1.1
from tensorflow.
Oh.. thanks, wish I'd read this sooner. I just finished installing Python 3.7 so I could avoid conda and get venv+pip working! LOL. Will try 3.8 instead now.
from tensorflow.
Hello,
I am seeing this same issue but neither python 3.8 nor the --extra-index flag seems to fix the problem. The pyindex command worked with no issues. Here is my current conda list for reference:
# Name Version Build Channel
ca-certificates 2021.10.8 h5b45459_0 conda-forge
nvidia-pyindex 1.0.9 pypi_0 pypi
openssl 3.0.0 h8ffe710_2 conda-forge
pip 22.0.4 pyhd8ed1ab_0 conda-forge
python 3.8.12 h900ac77_2_cpython conda-forge
python_abi 3.8 2_cp38 conda-forge
setuptools 60.9.3 py38haa244fe_0 conda-forge
sqlite 3.37.0 h8ffe710_0 conda-forge
ucrt 10.0.20348.0 h57928b3_0 conda-forge
vc 14.2 hb210afc_6 conda-forge
vs2015_runtime 14.29.30037 h902a5da_6 conda-forge
wheel 0.37.1 pyhd8ed1ab_0 conda-forge
from tensorflow.
@mrjoness we do not support conda
from tensorflow.
Sorry I should clarify that I am using the pip install, this was just to show package versions. This is the full error output:
C:\Users\mrjon\Projects\LSS>pip install --extra-index-url=https://pypi.ngc.nvidia.com --trusted-host pypi.ngc.nvidia.com nvidia-tensorflow[horovod]
Looking in indexes: https://pypi.org/simple, https://pypi.ngc.nvidia.com, https://pypi.ngc.nvidia.com
Collecting nvidia-tensorflow[horovod]
Downloading nvidia-tensorflow-0.0.1.dev5.tar.gz (7.9 kB)
Preparing metadata (setup.py) ... error
error: subprocess-exited-with-error
× python setup.py egg_info did not run successfully.
│ exit code: 1
╰─> [19 lines of output]
Traceback (most recent call last):
File "<string>", line 2, in <module>
File "<pip-setuptools-caller>", line 34, in <module>
File "C:\Users\mrjon\AppData\Local\Temp\pip-install-b_g8l9g9\nvidia-tensorflow_be127eece9114603ae5b578b2891349a\setup.py", line 150, in <module>
raise RuntimeError(open("ERROR.txt", "r").read())
RuntimeError:
###########################################################################################
The package you are trying to install is only a placeholder project on PyPI.org repository.
This package is hosted on NVIDIA Python Package Index.
This package can be installed as:
```
$ pip install nvidia-pyindex
$ pip install nvidia-tensorflow
```
Please refer to NVIDIA instructions: https://github.com/NVIDIA/tensorflow#install.
###########################################################################################
[end of output]
note: This error originates from a subprocess, and is likely not a problem with pip.
error: metadata-generation-failed
× Encountered error while generating package metadata.
╰─> See above for output.
note: This is an issue with the package mentioned above, not pip.
hint: See above for details.
from tensorflow.
I have the same issue under Ubuntu 22.04.1 LTS with Python 3.10.6
Trying:
pip install nvidia-pyindex pip install nvidia-tensorflow[horovod]
or
pip install --extra-index-url=https://pypi.ngc.nvidia.com --trusted-host pypi.ngc.nvidia.com nvidia-tensorflow
Both return the error previously shown which advises the package is a pypi.org placeholder and to see the install instructions.
EDIT: I notice that https://pypi.org/simple returns HTML that appears to be a list of packages, whereas https://pypi.ngc.nvidia.com redirects to https://developer.download.nvidia.com/compute/redist/ (which is a blank page)
I'm unsure whether the pypi.org HTML is there simply for the benefit of humans, or that's the actual index data that pip uses. If it's the latter, that may explain why pip cannot find nvidia-tensorflow on pypi.ngc.nvidia.com.
Facing the exact same issue. Any luck getting it to work?
from tensorflow.
In my case, I need to upgrade setuptools
since it used the system version, not the virtual environment version.
pip install -U setuptools
Then, I am able to install nvidia-tensorflow[horovod]
and nvidia-pyindex
without problem.
from tensorflow.
Related Issues (20)
- is tensorflow can be used with RTX A4000 HOT 1
- Wheel installation for Python 3.9 or above?
- How to Obtain libtensorflow_cc.lib in MSYS2 MinGW64 Environment with Bazel
- Install bug nvidia-tensorflow-0.0.1.dev5
- Incomplete bazel build command
- Is this fork still maintained by NVIDIA?
- Cannot dlopen some GPU libraries.
- NEM VOU POR LEGENDA.
- ANTI-FLOOD SYSTEM USING ARTIFICIAL INTELLIGENCE LILITI STK I.A 3.6.9
- pip >= 20.3 required, not 19.0 HOT 1
- tensorflow.python.framework.errors_impl.InternalError: cudaGetDevice() failed. Status: CUDA driver version is insufficient for CUDA runtime version HOT 4
- Building CUDA 12.1 (nv23.02) from source fails HOT 5
- latest release on nvidia-pyindex is broken HOT 4
- cannot install via wheels on nvidia/cuda HOT 2
- ERROR: Could not find a version that satisfies the requirement nvidia-tensorflow[horovod] (from versions: 0.0.1.dev4, 0.0.1.dev5) ERROR: No matching distribution found for nvidia-tensorflow[horovod] HOT 1
- Non-OK-status: GpuLaunchKernel error during distributed training of a large model
- cuda and tensorflow 1.15 are not compatible HOT 10
- Is TensorFlow can be used with RTX 4090? HOT 3
- nvidia-pyindex breaks pip HOT 5
- installation failure HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from tensorflow.