https://github.com/kurtamohler/pytorch-dataloader-segfault-investigation

Environments to help reproduct DataLoader segfault issue
https://github.com/kurtamohler/pytorch-dataloader-segfault-investigation

Last synced: 5 days ago
JSON representation

Environments to help reproduct DataLoader segfault issue

• Host: GitHub
• URL: https://github.com/kurtamohler/pytorch-dataloader-segfault-investigation
• Owner: kurtamohler
• Created: 2022-02-09T21:09:31.000Z (almost 3 years ago)
• Default Branch: master
• Last Pushed: 2022-02-09T21:09:54.000Z (almost 3 years ago)
• Last Synced: 2024-12-15T09:10:32.255Z (21 days ago)
• Language: Python
• Size: 2.93 KB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
This repo was created to help me attempt to reproduce an issue in PyTorch:

[link to issue](https://github.com/pytorch/pytorch/issues/31758)

## Prerequisite: Conda

Install Miniconda3: [link](https://docs.conda.io/en/latest/miniconda.html)

## Prerequisite: CUDA toolkits

The different environments in this repo require different CUDA versions. I used

these instructions to install multiple versions CUDA versions on my system:

[link](https://github.com/Quansight/dev-notes/blob/master/CUDA-installation-in-qgpu.md#installing-cuda-toolkits)

When creating an environment, if the required CUDA version is not found, an

error is thrown.

## Creating environments

Each of the environments in this repo have their own directory. They all

contain a `create-conda-env` script which creates the conda environment, sets

up variables to point to the proper CUDA version, and installs the corresponding

PyTorch version.

To create, for instance, env0, run the following:

```bash

$ ./env0/create-conda-env

```

This will create the conda environment `pytorch-dataloader-env0`. To activate it,

run:

```bash

$ conda activate pytorch-dataloader-env0

```

## Running the DataLoader segfault test

To run the DataLoader test, make sure one of the environments has been activated

and then run:

```bash

$ python test.py

```