Classify Traffic signs using a Convolutional Neural Network (a slightly modified LeNet architecture). We will train a model to classify traffic signs from the German Traffic Sign Dataset.
What is interesting here is that the training data are initially insufficient and we augment the whole database using different technics. For more details check the source code
To create an Anaconda environment on Ubuntu, run the following command in the terminal from the root folder of the project:
conda env create -f environment_ubuntu.yaml
Note On MacOs use environment_macos.yaml.