Shulou(Shulou.com)06/03 Report--

Xiaobian to share with you Win10 operating system PyTorch virtual environment configuration and PyCharm configuration example analysis, I believe most people do not know how, so share this article for your reference, I hope you read this article after a great harvest, let us go to understand it!

The details are as follows:

trial and error experience

First try out PyTorch version 1.0:

pip install https://download.pytorch.org/whl/cpu/torch-1.0.0-cp36-cp36m-win_amd64.whl

However, downloading these very slow, more than 70 megabytes of packets in China takes about half an hour, and a long time may cause timeout:

Some students use Tsinghua source/Keda source to accelerate, but I tried it and still couldn't o(￣ la ￣*), and finally adopted this method. After three overtime runs, finally... (It turns out that overtime is also a chance)

However, when installing torchvision, there was an error.,, I installed it for an afternoon and my face finally turned white once. You told me that there was no matching version o(<$￣*)

After that, he finally went on the right path.

installed version

Python3.6

conda4.3.8

torch0.4.0

torchversion0.4.1

installation process

1, create pytorch46 virtual environment (since python is 3.6, it is recorded as 36 here)

pytorch-cpu can set its own name, here I use pytorch46

conda create -n pytorch-cpu python=3.6

So even if it's created successfully

2. Look at the virtual environment you have created

conda info --envs

Those with an asterisk * are the currently active environments.

3. Activate the environment

Note that the following operations are operating in the pytorch46 virtual environment!

activate pytorch46

4, upgrade pip

Not surprisingly, the pip version here will be relatively low, so avoid future errors, or upgrade in advance worry

python -m pip install --upgrade pip

5, install torch-0.4.0 (CPU version)

Looking up online resources, we found that version 0.4 can already meet most of the requirements, and most of the code on GitHub is based on version 0.4 torch.

pip install http://download.pytorch.org/whl/cpu/torch-0.4.0-cp36-cp36m-win_amd64.whl

As you can see, the earlier versions were smaller, and the speed was acceptable even when downloaded from the original site.

6, install torchvision 0.4.1

pip install torchvision==0.4.1

7, test whether the installation is successful

import torchprint(torch)

As follows, you can successfully import torch and output it to show that the installation is successful.

8. Apply virtual environment pytorch46 to PyCharm project

If there is none in show all, click the + sign on the right to manually add

Find the installation directory of your virtual environment (if you can't find it, you can search the environment name you created directly in my computer)

After selecting, OK all the way, wait a moment, you can see that the environment originally created appears here

Select OK/Apply

9, testing.

Create test.py file

import torchprint(torch)x1 = torch.Tensor([3, 4])x2 = torch.FloatTensor([1, 2])print(x1)print(x2)

operation results

The above is "PyTorch virtual environment configuration and PyCharm configuration example analysis in Win10 operating system" All the contents of this article, thank you for reading! I believe that everyone has a certain understanding, hope to share the content to help everyone, if you still want to learn more knowledge, welcome to pay attention to the industry information channel!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.