1. ✏️ Setup
2. 💁 More infos
3. 🧪 Testing
4. ⚖️ License
5. 🔄 Changelog
6. 🐛 Bugs & TODO

You first need to build and compile the source code for your system (please replace x with the specific version numbers) :

pip uninstall tf_gan -y
python setup.py sdist bdist_wheel
pip install .\dist\tf_gan-x.x.x-py3-none-any.whl

Then import the library as follow :

import tf_gan as tfg

This hole program is supposed to run on python>=3.10.4.

Tensorflow >=2 is compatible with NVidia GPUs and requires almost no change in the code base (not required)

This is the page you are looking for. TL;DR :

• 450.80.02 minimum graphics drivers
• CUDA Toolkit
• NVIDIA cuDNN (please refer to this guide)

Then setup PATH variables, were x is the minor of the CUDA toolkit you have installed (make sure the paths are correct) :

SET PATH=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.x\bin;%PATH%
SET PATH=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.x\extras\CUPTI\lib64;%PATH%
SET PATH=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.x\include;%PATH%
SET PATH=C:\tools\cuda\bin;%PATH%

Also make sure the compute capability of the GPU isn't slowing down the hole process due to the (non)atomicity of some operations. Furthermore, using GPU isn't going to be particularly helpfull with a small number of features to train. Benefits will increase as the number of features to train goes past 4000 or so.

To begin, please make sure you have the necessary library up and ready on your environment :

pip install -r .\requirements.txt

Then run a batch of tests with unittest with :

python setup.py test

In addition, you can find code examples in the examples folder :

1. Add the examples directory to your PYTHONPATH environment variable with

   export PYTHONPATH=${TFGAN_REPO}/tf_gan/examples:${PYTHONPATH}

   Be sure to use the location where you cloned this repository.

2. Add this repository to your PYTHONPATH environment variable so that it can be used for tf_gan instead of any older libraries you might have installed.

   export PYTHONPATH=${TFGAN_REPO}:${PYTHONPATH}

3. Then navigate to the examples folder and run the training script.

   cd tf_gan/examples/
   python ./${EXAMPLE_NAME}/train.py

This project is licensed under the GPL-3.0 new or revised license. Please read the LICENSE file.

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

• Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
• Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
• Neither the name of the tf-gan authors nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

To view the full history, please read the full changelog. Latests changes :

gantt
    title Main Versions
    dateFormat YYYY-MM-DD

    section source Code (v0)
    v0.1 : 2022-06-23, 3d
    v0.2 :             1d
    v0.3 :             2d

    section source Code (v1)
    v1.0 : 2022-06-25, 2d
    v1.1 :             3d

    section source Code (v2)
    v2.0 : 2022-06-27, 1d
    v2.1 :             1d
    v2.2 :             2d
    v2.3 :             1d

    section Production release
    PyPI : 2022-07-01, 0d

bugs (final correction patch version)

• deprecated packages in imported libs : to be removed in python 3.12 and pillow 10
• tensorflow warnings about deleted checkpoint with unrestored values when not saving final run as gif or not running the "interactive" mode
• unable to decode temp data on examples

todo (first implementation version)

• provide examples, a lot of them
• push on PyPI