This model uses deep neural networks and convolutional neural networks to classify traffic signs. The model will be trained on traffic signs from the German Traffic Sign Dataset. A training set, validation set and testing set were used. The testing set was keeped seperate and untouched until after the model was completed. The model had an accuracy of 94% on the testing set.
The report can be viewed by clicking the notebook Traffic_Signs_Recognition.ipynb above.
Also, later after I learnd Keras I rewrote the model with Keras code. This model can be viewed by clicking traffic-sign-classification-with-keras.ipynb above. The test accuracy increased to 95.9%