Training FuseNet in a Docker Container
FuseNet is developed as a general architecture for deep convolutional neural network (CNN) to train dataset with RGB-D images. It can be used for semantic segmentation, scene classification and other applications. The FuseNet architecture has been implemented in Caffe and Pytorch before. However, creating an appropriate pipeline for training FuseNet in a docker container can be of much help by reducing the need to installing and managing the required packages on every new machine. This repository has gathered all the essential steps for training the FuseNet in a docker file. The same approach can be used for dockerizing other neural network models as well.