Traffic - CS50 AI With Python

By Roi Solomon

I used the The German Traffic Sign Recognition Benchmark (GTSRB) database.

What I Found?

Tested the effects of hidden layers on the accuracy of the model.

Hidden Layers:

        # Another hidden layer
        tf.keras.layers.Dense(NUM_CATEGORIES * 16, activation="relu"),

        # Another hidden layer
        tf.keras.layers.Dense(NUM_CATEGORIES * 8, activation="relu"),

        # Another hidden layer
        tf.keras.layers.Dense(NUM_CATEGORIES * 4, activation="relu"),

Result - 97% Acccuracy:

GTSRB Database:

The database contains 43 subdirectories in this dataset, numbered 0 through 42. Each numbered subdirectory represents a different category (a different type of road sign). Within each traffic sign’s directory is a collection of images of that type of traffic sign.

Download

Sources

• TensorFlow The Sequential model - Link
• Loading Custom Image Dataset for Deep Learning Models: Part 1 - Link
• Understand the Softmax Function in Minutes - Link
• CS50 Lecture 5 Notes - Link

License

MIT