Installation and Setup 🛠️

Installation 📥

Pyro-NN works with PyTorch and TensorFlow. This guide focuses on PyTorch.

Requirements:

Microsoft Visual Studio (for self-build) 💻
Microsoft Visual C++ 14.0+
Python packagebuild🐍
CUDA >10.2 🚀

Once all of these requirements are fulfilled, follow these steps::

Clone the repo and switch to the develop branch if not done yet:
```
git checkout develop
```

Modify TOML file if needed; please make sure the torch version in TOML file is the same as your environment, or DLL errors will occur:

 requires = ["setuptools==69.5.1",
           "ninja",
           "tensorflow[and-cuda]==2.11.1",
           "--extra-index-url https://download.pytorch.org/whl/cu118",
           "torch==2.3.0+cu118",
   ]

Run the command:
```
python -m build . 
```
Switch to the newly created sub-directory dist
```
cd dist/ 
```
Build the wheel file using the following command:
```
pip install pyronn-(*version_number*).whl
```

Note

    If necessary, you can modify the `pyproject.toml` file to specify a 
    particular torch version. However, be cautious and only make changes 
    if you are confident in what you are doing!

If you encounter a problem during installation, have a look at our wiki: https://github.com/csyben/PYRO-NN/wiki