StanleyTheBot

intro

StanleyTheBot is a simple Python web-app that serves as a image recognition chat-bot. The app simply asks for the name of the user, then asks the user for a photo to classify.

background

I've always had great interest in Artificial Intelligence and I have strive to thrive in this field. In my endeavor to understand A.I., I realized that Python is one of the major programming languages in the field. So, I decided to learn Python >at least the basics. After getting a comfortable with the basics of Python, I wanted to build a simple program with Python that relates to A.I. Thereby, I created StanleyTheBot.

backbone

The base of StanleyTheBot is Flask and Jinja. Flask framework handles the dynamicity of the app: redirection of pages, variables in HTML. Images are classified with IBM Watson Visual Recognition API. Firebase storage is used to store the images uploaded. UI is done with MaterializeCSS. StanleyTheBot is deployed to Heroku StanleyTheBot.herokuapp.com.

walkthrough

The main python file that controls the app is index.py. All the web pages in this app are in templates folder. Static folder holds other files that are used in the HTML pages. The following files are used for deploying the app to Heroku:

• _pycache_
• config.py
• config.pyc
• Procfile
• requirements.txt
• settings.py

demo

1. Stanley introduces itself and asks for user's name.
2. User provides a non-empty name.
3. Stanley mentions its purpose.
4. User picks a photo.
5. Stanley uploads the photo to Firebase storage.
6. Link to photo is received from Firebase.
7. IBM Watson uses the link to access photo.
8. IBM Watson returns a dictionary.
9. Stanley processes the dictionary.
10. Stanley replies user with processed result.