A simple LeNet based classifier used to classify German Traffic Signs Dataset.
The model for training the dataset involves some adaptations from the research paper titled Traffic Sign Recognition with Multi-Scale Convolutional Networks by Pierre Sermanet and Yann LeCun.
- Traffic_Sign_Classifier.ipynb
- The file consists of all the required cells to load, preprocess, train, validate and test the data.
- net.jpg
- It is a reference image for the architecture of the classifier model deployed in the project.
- /extra_images
- Consists of some random images downloaded from internet to test the classifier
- Anaconda
- Download the appropriate Anaconda version as per your local system.
- Download the German Traffic Signs Dataset. You can also use any other appropriate dataset that can be compatible with the model.
- Using Jupyter Notebook open the .ipynb file
- Execute the cells.
- Enjoy the "World of Deep Learning" ;)
- Included in the repo