Pytorch ошибка при установке

So you have Cuda 10 installed? If you do, try this:

pip3 install https://download.pytorch.org/whl/cu100/torch-1.2.0-cp37-cp37m-win_amd64.whl

followed by:

pip3 install torchvision

To check if it was installed properly, type this into your command line:

python

followed by:

from __future__ import print_function
import torch
x = torch.rand(5, 3)
print(x)

If you get this output:

tensor([[0.3380, 0.3845, 0.3217],
        [0.8337, 0.9050, 0.2650],
        [0.2979, 0.7141, 0.9069],
        [0.1449, 0.1132, 0.1375],
        [0.4675, 0.3947, 0.1426]])

PyTorch was installed correctly!

I am trying to install pytorch with pip using

pip install torch

or

pip3 install torch===1.3.1 torchvision===0.4.2 -f https://download.pytorch.org/whl/torch_stable.html

with python 3.7.4

and with python 3.8 (latest stable release)

both on 32 and 64 bit.

and getting

Collecting torch   Using cached
https://files.pythonhosted.org/packages/f8/02/880b468bd382dc79896eaecbeb8ce95e9c4b99a24902874a2cef0b562cea/torch-0.1.2.post2.tar.gz
Collecting pyyaml (from torch)   Downloading
https://files.pythonhosted.org/packages/bc/3f/4f733cd0b1b675f34beb290d465a65e0f06b492c00b111d1b75125062de1/PyYAML-5.1.2-cp37-cp37m-win_amd64.whl
(215kB)
    100% |████████████████████████████████| 225kB 1.2MB/s Installing collected packages: pyyaml, torch   Running setup.py install for torch
... error
    Complete output from command C:NoamCodethreadartstav-rlvenvScriptspython.exe -u -c "import
setuptools,
tokenize;__file__='C:\Users\noams\AppData\Local\Temp\pip-install-djc6s2t8\torch\setup.py';f=getattr(tokenize,
'open', open)(__file__);code=f.read( ).replace('rn',
'n');f.close();exec(compile(code, __file__, 'exec'))" install
--record C:UsersnoamsAppDataLocalTemppip-record-zohv2zo7install-record.txt
--single-version-externally-managed --compile --install-headers C:NoamCodethreadartstav-rlvenvinclu desitepython3.7torch:
    running install
    running build_deps
    Traceback (most recent call last):
      File "<string>", line 1, in <module>
      File "C:UsersnoamsAppDataLocalTemppip-install-djc6s2t8torchsetup.py",
line 265, in <module>
        description="Tensors and Dynamic neural networks in Python with strong GPU acceleration",
      File "C:NoamCodethreadartstav-rlvenvlibsite-packagessetuptools-40.8.0-py3.7.eggsetuptools__init__.py",
line 145, in setup
      File "C:Python37_x64libdistutilscore.py", line 148, in setup
        dist.run_commands()
      File "C:Python37_x64libdistutilsdist.py", line 966, in run_commands
        self.run_command(cmd)
      File "C:Python37_x64libdistutilsdist.py", line 985, in run_command
        cmd_obj.run()
      File "C:UsersnoamsAppDataLocalTemppip-install-djc6s2t8torchsetup.py",
line 99, in run
        self.run_command('build_deps')
      File "C:Python37_x64libdistutilscmd.py", line 313, in run_command
        self.distribution.run_command(command)
      File "C:Python37_x64libdistutilsdist.py", line 985, in run_command
        cmd_obj.run()
      File "C:UsersnoamsAppDataLocalTemppip-install-djc6s2t8torchsetup.py",
line 51, in run
        from tools.nnwrap import generate_wrappers as generate_nn_wrappers
    ModuleNotFoundError: No module named 'tools.nnwrap'

    ---------------------------------------- Command "C:NoamCodethreadartstav-rlvenvScriptspython.exe -u -c "import
setuptools,
tokenize;__file__='C:\Users\noams\AppData\Local\Temp\pip-install-djc6s2t8\torch\setup.py';f=getattr(tokenize,
'open', open)(__file__);code=f.read().replace('rn', 'n');
f.close();exec(compile(code, __file__, 'exec'))" install --record
C:UsersnoamsAppDataLocalTemppip-record-zohv2zo7install-record.txt
--single-version-externally-managed --compile --install-headers C:NoamCodethreadartstav-rlvenvincludesitepython3.7torch" 
failed with error code 1 in
C:UsersnoamsAppDataLocalTemppip-install-djc6s2t8torch

clearly, I am doing something wrong.

Please help!

I’m also getting the same problem, trying to install pytorch in python3.9 with macOS Catalina if that’s relevant.

What’s the ‘current’ download link for a stable version?

(Error prints are:

Building wheel for torch (setup.py) … error
ERROR: Command errored out with exit status 1:
command: /usr/local/opt/python@3.9/bin/python3.9 -u -c ‘import sys, setuptools, tokenize; sys.argv[0] = ‘»‘»‘/private/var/folders/tb/p0rc6qfs4tl3q88qqn0d1dth0000gn/T/pip-install-5_srv_0q/torch/setup.py'»‘»‘; file='»‘»‘/private/var/folders/tb/p0rc6qfs4tl3q88qqn0d1dth0000gn/T/pip-install-5_srv_0q/torch/setup.py'»‘»‘;f=getattr(tokenize, ‘»‘»‘open'»‘»‘, open)(file);code=f.read().replace(‘»‘»‘rn'»‘»‘, ‘»‘»‘n'»‘»‘);f.close();exec(compile(code, file, ‘»‘»‘exec'»‘»‘))’ bdist_wheel -d /private/var/folders/tb/p0rc6qfs4tl3q88qqn0d1dth0000gn/T/pip-wheel-g_u1i7cz
cwd: /private/var/folders/tb/p0rc6qfs4tl3q88qqn0d1dth0000gn/T/pip-install-5_srv_0q/torch/
Complete output (30 lines):
running bdist_wheel
running build
running build_deps
Traceback (most recent call last):
File «», line 1, in
File «/private/var/folders/tb/p0rc6qfs4tl3q88qqn0d1dth0000gn/T/pip-install-5_srv_0q/torch/setup.py», line 225, in
setup(name=»torch», version=»0.1.2.post2″,
File «/usr/local/lib/python3.9/site-packages/setuptools/init.py», line 153, in setup
return distutils.core.setup(**attrs)
File «/usr/local/Cellar/python@3.9/3.9.0_1/Frameworks/Python.framework/Versions/3.9/lib/python3.9/distutils/core.py», line 148, in setup
dist.run_commands()
File «/usr/local/Cellar/python@3.9/3.9.0_1/Frameworks/Python.framework/Versions/3.9/lib/python3.9/distutils/dist.py», line 966, in run_commands
self.run_command(cmd)
File «/usr/local/Cellar/python@3.9/3.9.0_1/Frameworks/Python.framework/Versions/3.9/lib/python3.9/distutils/dist.py», line 985, in run_command
cmd_obj.run()
File «/usr/local/lib/python3.9/site-packages/wheel/bdist_wheel.py», line 290, in run
self.run_command(‘build’)
File «/usr/local/Cellar/python@3.9/3.9.0_1/Frameworks/Python.framework/Versions/3.9/lib/python3.9/distutils/cmd.py», line 313, in run_command
self.distribution.run_command(command)
File «/usr/local/Cellar/python@3.9/3.9.0_1/Frameworks/Python.framework/Versions/3.9/lib/python3.9/distutils/dist.py», line 985, in run_command
cmd_obj.run()
File «/usr/local/Cellar/python@3.9/3.9.0_1/Frameworks/Python.framework/Versions/3.9/lib/python3.9/distutils/command/build.py», line 135, in run
self.run_command(cmd_name)
File «/usr/local/Cellar/python@3.9/3.9.0_1/Frameworks/Python.framework/Versions/3.9/lib/python3.9/distutils/cmd.py», line 313, in run_command
self.distribution.run_command(command)
File «/usr/local/Cellar/python@3.9/3.9.0_1/Frameworks/Python.framework/Versions/3.9/lib/python3.9/distutils/dist.py», line 985, in run_command
cmd_obj.run()
File «/private/var/folders/tb/p0rc6qfs4tl3q88qqn0d1dth0000gn/T/pip-install-5_srv_0q/torch/setup.py», line 51, in run
from tools.nnwrap import generate_wrappers as generate_nn_wrappers
ModuleNotFoundError: No module named ‘tools.nnwrap’

ERROR: Failed building wheel for torch
Running setup.py clean for torch
ERROR: Command errored out with exit status 1:
command: /usr/local/opt/python@3.9/bin/python3.9 -u -c ‘import sys, setuptools, tokenize; sys.argv[0] = ‘»‘»‘/private/var/folders/tb/p0rc6qfs4tl3q88qqn0d1dth0000gn/T/pip-install-5_srv_0q/torch/setup.py'»‘»‘; file='»‘»‘/private/var/folders/tb/p0rc6qfs4tl3q88qqn0d1dth0000gn/T/pip-install-5_srv_0q/torch/setup.py'»‘»‘;f=getattr(tokenize, ‘»‘»‘open'»‘»‘, open)(file);code=f.read().replace(‘»‘»‘rn'»‘»‘, ‘»‘»‘n'»‘»‘);f.close();exec(compile(code, file, ‘»‘»‘exec'»‘»‘))’ clean —all
cwd: /private/var/folders/tb/p0rc6qfs4tl3q88qqn0d1dth0000gn/T/pip-install-5_srv_0q/torch
Complete output (2 lines):
running clean
error: [Errno 2] No such file or directory: ‘.gitignore’

ERROR: Failed cleaning build dir for torch
Building wheel for pyyaml (setup.py) … done
Created wheel for pyyaml: filename=PyYAML-5.3.1-cp39-cp39-macosx_10_15_x86_64.whl size=44623 sha256=27a191c5d9df245e8d5ca7764ad4e2ba8fd1e7acfa26bb8501c6b96f1c27e6d6
Stored in directory: /Users/Gail/Library/Caches/pip/wheels/69/60/81/5cd74b8ee068fbe9e04ca0d53148f28f5c6e2c5b177d5dd622
Successfully built pyyaml
Failed to build torch
Installing collected packages: pyyaml, torch
Running setup.py install for torch … error
ERROR: Command errored out with exit status 1:
command: /usr/local/opt/python@3.9/bin/python3.9 -u -c ‘import sys, setuptools, tokenize; sys.argv[0] = ‘»‘»‘/private/var/folders/tb/p0rc6qfs4tl3q88qqn0d1dth0000gn/T/pip-install-5_srv_0q/torch/setup.py'»‘»‘; file='»‘»‘/private/var/folders/tb/p0rc6qfs4tl3q88qqn0d1dth0000gn/T/pip-install-5_srv_0q/torch/setup.py'»‘»‘;f=getattr(tokenize, ‘»‘»‘open'»‘»‘, open)(file);code=f.read().replace(‘»‘»‘rn'»‘»‘, ‘»‘»‘n'»‘»‘);f.close();exec(compile(code, file, ‘»‘»‘exec'»‘»‘))’ install —record /private/var/folders/tb/p0rc6qfs4tl3q88qqn0d1dth0000gn/T/pip-record-qwqhjxg9/install-record.txt —single-version-externally-managed —compile —install-headers /usr/local/include/python3.9/torch
cwd: /private/var/folders/tb/p0rc6qfs4tl3q88qqn0d1dth0000gn/T/pip-install-5_srv_0q/torch/
Complete output (23 lines):
running install
running build_deps
Traceback (most recent call last):
File «», line 1, in
File «/private/var/folders/tb/p0rc6qfs4tl3q88qqn0d1dth0000gn/T/pip-install-5_srv_0q/torch/setup.py», line 225, in
setup(name=»torch», version=»0.1.2.post2″,
File «/usr/local/lib/python3.9/site-packages/setuptools/init.py», line 153, in setup
return distutils.core.setup(**attrs)
File «/usr/local/Cellar/python@3.9/3.9.0_1/Frameworks/Python.framework/Versions/3.9/lib/python3.9/distutils/core.py», line 148, in setup
dist.run_commands()
File «/usr/local/Cellar/python@3.9/3.9.0_1/Frameworks/Python.framework/Versions/3.9/lib/python3.9/distutils/dist.py», line 966, in run_commands
self.run_command(cmd)
File «/usr/local/Cellar/python@3.9/3.9.0_1/Frameworks/Python.framework/Versions/3.9/lib/python3.9/distutils/dist.py», line 985, in run_command
cmd_obj.run()
File «/private/var/folders/tb/p0rc6qfs4tl3q88qqn0d1dth0000gn/T/pip-install-5_srv_0q/torch/setup.py», line 99, in run
self.run_command(‘build_deps’)
File «/usr/local/Cellar/python@3.9/3.9.0_1/Frameworks/Python.framework/Versions/3.9/lib/python3.9/distutils/cmd.py», line 313, in run_command
self.distribution.run_command(command)
File «/usr/local/Cellar/python@3.9/3.9.0_1/Frameworks/Python.framework/Versions/3.9/lib/python3.9/distutils/dist.py», line 985, in run_command
cmd_obj.run()
File «/private/var/folders/tb/p0rc6qfs4tl3q88qqn0d1dth0000gn/T/pip-install-5_srv_0q/torch/setup.py», line 51, in run
from tools.nnwrap import generate_wrappers as generate_nn_wrappers
ModuleNotFoundError: No module named ‘tools.nnwrap’

ERROR: Command errored out with exit status 1: /usr/local/opt/python@3.9/bin/python3.9 -u -c ‘import sys, setuptools, tokenize; sys.argv[0] = ‘»‘»‘/private/var/folders/tb/p0rc6qfs4tl3q88qqn0d1dth0000gn/T/pip-install-5_srv_0q/torch/setup.py'»‘»‘; file='»‘»‘/private/var/folders/tb/p0rc6qfs4tl3q88qqn0d1dth0000gn/T/pip-install-5_srv_0q/torch/setup.py'»‘»‘;f=getattr(tokenize, ‘»‘»‘open'»‘»‘, open)(file);code=f.read().replace(‘»‘»‘rn'»‘»‘, ‘»‘»‘n'»‘»‘);f.close();exec(compile(code, file, ‘»‘»‘exec'»‘»‘))’ install —record /private/var/folders/tb/p0rc6qfs4tl3q88qqn0d1dth0000gn/T/pip-record-qwqhjxg9/install-record.txt —single-version-externally-managed —compile —install-headers /usr/local/include/python3.9/torch Check the logs for full command output.

The RuntimeError: Couldn’t Install ‘torch’ error occurs when there is a problem installing the ‘torch’ library, which is required for PyTorch, a popular machine learning library.

Causes of the RuntimeError

1. You have insufficient permissions to install the library.
2. It conflicts with an existing installation of ‘torch’ or other dependencies.
3. It is an incompatible system or environment.
4. Network connectivity issues.
5. The current Python version is not compatible with the ‘torch’.

Reproduce the error

import sys
import os

# Check if user is root
if not os.geteuid() == 0:
  sys.exit("nYou must be root to run this script.n")

os.system("pip install pytorch")
# Check if installation was successful
import torch

# If the import fails, raise a runtime error
if not torch:
  raise RuntimeError("Couldn't Install 'torch'")

The above code checks if the script’s user is root and then attempts to install PyTorch using pip.

If the installation fails, it raises a RuntimeError with the message “Couldn’t Install ‘torch’”.

You can use one of the following fixes to solve the “RuntimeError: Couldn’t Install ‘torch’” error.

1. You must check that you have the necessary permissions to install the library.
2. Make sure that you have the latest version of pip and setuptools.
3. Uninstall any conflicting versions of the ‘torch’ library or other dependencies.
4. Ensure that your system and environment meet the requirements for PyTorch, including the correct version of Python.
5. You need to check your network connectivity.

I hope one of the above solutions works for you.

Usually, we often run into errors like “runtimeerror: couldn’t install torch.”. It is one of the most common errors that developers may encounter during running their code.

The “Runtimeerror: couldnt install torch” error typically occurs because there is a problem with installing the PyTorch library on your system.

We will explain to you in detail to understand this error, what it means, and how we can fix it.

Why does this error couldn’t install torch occur?

The error “Couldn’t install torch” typically occurs when there is an issue with the installation process of the PyTorch library.

Also, read the articles below to understand more about runtimeerror:

Runtimeerror: cuda error: device-side assert triggered

How to Fix this Error?

Here are the step-by-step solutions to help you to fix the “Runtimeerror: couldnt install torch” error:

Step 1: Check your Python version

The PyTorch requires a Python 3.5 or higher. Make sure you have the correct version of Python installed on your system.

You can check your Python version by running the following command in your terminal or command prompt:

python --version

Step 2: Update pip and setuptools

You will make sure that you have the latest versions of pip and setuptools installed on your system. You can update them by running the following commands:

#UPGRADE PIP
pip install --upgrade pip

#UPGRADE SETUPTOOLS
pip install --upgrade setuptools

Step 3: Install PyTorch via pip

The simple and easiest way to install PyTorch is through pip package. You can install the CPU version of PyTorch by running the following command:

pip install torch

If you want to install the GPU version of PyTorch, you can do this by running the following command:

pip install torch torchvision torchaudio -f https://download.pytorch.org/whl/cu111/torch_stable.html

Step 4: Check your CUDA version

If you are installing the GPU version of PyTorch, make sure you have the correct version of CUDA installed on your system.

You can check your CUDA version by running the following command:

nvcc --version

Step 5: Check your cuDNN version

When you are installing the GPU version of PyTorch, ensure that you have the correct version of cuDNN installed on your system.

You can check your cuDNN version by running the following command:

cat /usr/local/cuda/include/cudnn.h | grep CUDNN_MAJOR -A 2

Step 6: Check your system requirements

You will make sure that your system meets the minimum requirements for PyTorch. You can find the system requirements on the PyTorch documentation on their website.

Step 7: Create a virtual environment

Creating a virtual environment is a good practice to prevent conflicts between packages. You can create a virtual environment by running the following command:

python -m venv env

Step 8: Install PyTorch in the virtual environment

Activate the virtual environment and install PyTorch by running the following commands:

source env/bin/activate
pip install torch

Step 9: Uninstall and reinstall PyTorch

If the above steps not working, you can try uninstalling PyTorch and reinstalling it. To uninstall PyTorch, run the following command:

#UNINSTALL
pip uninstall torch

Then, reinstall PyTorch by running one of the installation commands mentioned in Step 3 or Step 8, it depends on whether you want to install the CPU or GPU version.

If the PyTorch has been installed successfully, you can check the installation by opening a Python interpreter and running the following command:

import torch
print(torch.__version__)

This code above will print the version of PyTorch you just installed.

Conclusion

The “Runtimeerror: couldnt install torch” error can be frustrating, yet it is usually easy to fix. By following the steps in this article, you should be able to install PyTorch successfully.

We hope that this article has been helpful in resolving the error.

FAQs