pyvtorch 1.2.0

Python tools to work in deep learning with PyTorch

PyVTorch

General custom tools to work in deep learning research using Python and PyTorch

Getting Started

Installation

This package can easily be installed using pip:

pip install pyvtorch

An alternative installation that partially uses Anaconda would involve...

1. First, install some Anaconda distribution, in case you do not have any: https://docs.anaconda.com/anaconda/install/
2. Then, create an Anaconda environment with Python 3.11.0

   conda create -n dev python=3.11.0
   

3. Activate the environment

   conda activate dev
   

4. Then, install all required packages by running the install.sh script:

   yes | . install.sh
   

   This will have executed...

   conda install python=3.11.0 \
       pytorch::pytorch pytorch::torchvision pytorch::pytorch-cuda \
       numpy scikit-image scikit-learn conda-forge::matplotlib tqdm \
       -c nvidia
    pip install pyvtools pyvtorch
   

5. You can make sure that your PyTorch installation has CUDA GPU support by running...

   python -c "import torch; print(torch.cuda.is_available()) \
              print([torch.cuda.get_device_name(i) for i in range(torch.cuda.device_count())])"  
   

   The first line should print True if CUDA is supported. And the second line should show you the name/s of your available GPU/s.
6. That's it! You're good to go :)

That second installation procedure is designed to be overly redundant, so please feel free to follow your own installation procedure.

Requirements

Provided installation steps are only guaranteed to work in Ubuntu 24.04 with NVidia drivers 535.

In case you are following another installation procedure, this repository requires...

• Python 3.11.0
• PyVTools >= 1.2.0

Additional information

Main Author Contact

Valeria Pais - @vrpais - [e-mail]