Stock analysis frontend

Start locally

1. Clone this repo

   git clone https://github.com/xcollantes/stock-analysis-frontend

2. Download dependencies:

   python3 -m venv env
   env/bin/pip install -r requirements.txt

3. Create .streamlit/secrets.toml file with API keys:

   • Get FinancialModelPrep.com API key: https://site.financialmodelingprep.com/register
   • Get FinnHub.io API key: https://finnhub.io

4. Start locally:

   env/bin/streamlit run 1_🏠_Getting_started.py

   Default is Home.py but to make path and sidebar named with emoji, this is the best way.

Adding a new page

The Home page is Home.py which is accessed at http://localhost.

Subsequent pages can be nested under the pages/ directory and accessed:

pages/drops.py -> http://localhost/drops

1. Add page as file in pages/.
2. Add page name and title to deps/Home.py.

Naming pages

File name will tell Streamlit:

• Order
• Emoji
• Sidebar title
• Url path

Example: pages/1_📈_Plotting_Demo.py

• Order: first in sidebar
• Emoji: chart emoji
• Sidebar title: Plotting Demo
• Url path: https://mystreamlit.streamlit.app/Plotting_Demo

Streamlit docs

Common debugging issues

Yahoo Finance.

yfinance API is the weakest part of this project because it is a free resource which scrapes Yahoo Finance data. Check functions that make contact with Yahoo Finance first.

#35

TypeError: 'DataFrame' object is not callable.

Error message is often misleading and usually is actually the failure of the yfinance API. Either update the library or wait 24 hours for any limits to replenish.

After checking the Yahoo Finance, check other keys for the other data sources.

Enter password showing up in the middle of the page.

The rest of the page could not load because the feature fell back on the password box.