jhoeller/computer-vision repository overview

⁠NVIDIA GPU/Tensor Core Accelerator for PyTorch, PyTorch Geometric, Tensorflow 2, Tensorboard + OpenCV

A complete computer vision container that includes Jupyter notebooks with built-in code hinting, Miniconda, CUDA 11, TensorRT inference accelerator for Tensor cores, CuPy (GPU drop in replacement for Numpy), PyTorch, PyTorch Geometric for geomteric learning and/or Graph Neural Networks, TF2, Tensorboard, and OpenCV (complied for CUDA) for accelerated workloads on NVIDIA Tensor cores and GPUs.

There are working notebook examples on how to wire up, both Torch and TF2 to Tensorboard in /app folder.

---

It takes a while for OpenCV to compile for CUDA, so if you want to skip the build process to save time you can run and pull the pre-built image with this command

docker run --rm -it --runtime=nvidia --user $(id -u):$(id -g) --group-add container_user --group-add sudo -v "${PWD}:/app" -p 8888:8888 -p 6006:6006 jhoeller/computer-vision:latest

---

⁠Features

• Miniconda: Accelerated Python, version 3.9
• CuPy: GPU accelerated drop in replacement for Numpy
• OpenCV, latest version which is made to compile for CUDA GPUs in the container (Depending upon your GPU you may have to change -DCUDA_ARCH_BIN=7.5 in the OpenCV flags within the Dockerfile.
• PyTorch with Torchvision for GPU, latest version
• PyTorch geometric for GNN's
• Captum to explain Torch models
• Tensorflow 2 with Keras
• Tensorboard for both Torch and TF2
• NVIDIA TensorRT inference accelerator for Tensor core access and CUDA 11 for GPUs
• Repo includes two working notebook examples on how to wire up Torch and TF2 to TensorBoard, located in /app folder