TensorFlow2 Model Garden document site and YOLO project tutorial

<p>I listed three potential projects, the main directions are implement MLCV research papers with GPU CUDA accelerating and distributing API, and code examples with tutorials on Colab. I would like to do any one direction (implement more) or two directions (which works can be supplement and adjusted models for more improvement). ; For CUDA C++ accelerating, my current ideas are concurrency improvement (on host) and parallel and shared memory improvement (on device), and I open for urgent project of Eigen's Module. For develop core code examples, my current ideas are implement the most popular models such as Mask R-CNN, Fast R-CNN, YOLO etc, include example code and visualization tool), and also interested in development the GPU accelerating distributing API. For the TF model garden, my current ideas are implement the models listed in the Project2 or other new research paper by TF on Colab notebook.</p>