P3 CB Writeup | Torin Wolff

Section 1, Collections + ML: The bert.py file in my Flask API uses a pre-trained TensorFlow model (cocomodel.h5) to interpret image data. This model is trained on the COCO dataset and processes the image through several layers to extract features and detect objects or dominant objects/shapes. The image is then preprocessed, passed through the model to generate a list of detected objects, and then post-processed into a human-readable form via Gemini from Google. The detected objects are then passed to a BERT model to generate a human-like text response. The API defines a route /genimage that accepts POST requests with an image file, processes the image with the model, and returns the generated text.
- Section 2, SQLite: In my program, I am able to go back to the Stocks search function where I allowed users to be able to write a missing stock in the case that they don't find the stock they want to search for, heres the code for user adding data to the SQLite db:
```
def put(self):
    try:
        body = request.get_json()
        symbol = body.get('symbol')
        company = body.get('company')
        quantity = body.get('quantity')
        sheesh = body.get('sheesh')
        ## Add new row to the sqlite db through the users.py MissingStock class
        print(symbol, company, quantity, sheesh)
        new_stock = Stocks(symbol, company, quantity, sheesh)  # Use Stocks instead of MissingStock
        db.session.add(new_stock)
        db.session.commit()
        response = {
            'message': f'{symbol} has been added to the database'
        }
        return response, 200
    except Exception as e:
        return {'error': str(e)}, 500
api.add_resource(stock, '/missingstock')
```
  The searchstocks.py file in the Flask application interacts with a SQLite database using SQLAlchemy. It defines a route /missingstock that accepts POST requests, creates a new Stocks record with the provided data, and adds it to the database. If there's an error, it catches the exception and returns an error message. The file also defines a search_stock route that accepts POST requests, retrieves the stock symbol from the request data, and makes an API call to the Financial Modeling Prep API to get information about the stock. If the API call is unsuccessful, it returns an error message.
- Section 1, List/Dictionary & Postman Heres the video for my postman demo as well as debugger: https://github.com/TDWolff/AtlasIndex/assets/142455935/b5f769d2-8882-40b2-9e22-d89568ba3182

Section 1, Frontend: Post Method: https://github.com/TDWolff/AtlasIndex/assets/142455935/717ca8f1-49a1-4164-9205-1ef8e998a019
Section 2, Frontend:

This is a picture of the stocks frontend code where it sends information as a put to the backend where it just says weather it worked or not.

In this picture, i made a post to the backend to send data and receive data back.

Overall my projects consisted of making two different modules, a stock search where i accessed a public library of stocks and pulled one based off its name and then if the user can't find a stock, they can add it, and my other feature was implementing an ai chatbot combined with image detection to achieve an advanced chatbot better than chatgpt, in my past time i have been working on a more advanced version that can use speech to text as well as making it a voice with text to speech and also spotify to allow it to play music based on commands.

TDWolff / AtlasIndex

P3 CB Writeup | Torin Wolff #17