How to Install and Run Ollama on macOS and Create a Stock Report with Python

This tutorial guides you through installing Ollama on macOS to run AI large language models locally, interacting with the Mistral model in the terminal, and creating a Python program to generate a daily stock report using live stock data. The program will save the report to a file, send it via Telegram, and be scheduled to run automatically using cron. This is for educational purposes only; I am not responsible for your stock trading decisions.

Prerequisites

• A macOS system with at least 16GB RAM (recommended for Mistral model).
• Visual Studio Code installed.
• Python 3 installed.
• A Telegram bot set up with a bot token and chat ID for sending messages (not covered in this tutorial; ensure you have a working telegram_message function).

Step 1: Install and Set Up Ollama

1. Download Ollama:

   • Visit the Ollama website and click "Download for macOS".
   • Open the downloaded file and drag the Ollama app to your Applications folder.

2. Start the Ollama Server:

   • Open Terminal and run:

     ollama serve
     

   • This starts the Ollama server in the background.

3. Download the Mistral Model:

   • Open a new Terminal window (Command + T).
   • Pull the Mistral model, which is lightweight and suitable for 16GB RAM systems:

     ollama pull mistral
     

   • Wait a few minutes for the download to complete, depending on your internet speed.

4. Run and Test the Mistral Model:

   • In the new Terminal window, start the Mistral model:

     ollama run mistral
     

   • Wait a few seconds for it to boot up.
   • Test it by typing questions, such as:

     Where is Badlands National Park?
     

     or

     Why is the sky blue?
     

   • The model runs locally on your Mac’s CPU without an internet connection.

Step 2: Set Up the Python Environment

1. Open Visual Studio Code:

   • Launch Visual Studio Code.

2. Create a Virtual Environment:

   • Open the integrated terminal in Visual Studio Code (View > Terminal or Ctrl + ~).
   • Create a virtual environment:

     python -m venv venv
     

   • Activate the virtual environment:

     source venv/bin/activate
     

   • You should see (venv) in the terminal prompt.

3. Install Required Libraries:

   • Install the yfinance and requests libraries:

     pip install yfinance requests
     

Step 3: Create the Stock Report Program

1. Create a New File:

   • In Visual Studio Code, go to File > New File or press Ctrl + N.
   • Save the file as stock_report.py (File > Save As or Ctrl + S).

2. Add the Stock Report Code:

   • Copy and paste the following code into stock_report.py:

     import yfinance as yf
     import requests
     import json
     from telegram import telegram_message
     
     # List of stocks
     stocks = ["AAPL", "MSFT", "NVDA", "AMZN", "TSLA"]
     
     # Get stock data for the past day
     data = yf.download(stocks, period="1d")
     
     # Create a prompt for Ollama with the stock data
     prompt = f"Generate a short report on these stocks for the past 24 hours: {', '.join(stocks)}. Data: {data.to_dict()}"
     print(f"Prompt: {prompt}")
     
     # Send the prompt to Ollama
     response = requests.post("http://localhost:11434/api/generate", json={"model": "mistral", "prompt": prompt, "stream": False})
     result = response.json()
     report = result.get("response", "")
     
     print('
     
     ')
     print('--------------------Report--------------------')
     
     # Display the report
     print(report)
     
     # Save the report to a text file
     with open('stock_report.txt', 'w') as f:
         f.write(report)
     
     # Send the report via Telegram
     telegram_message(report)

   • Save the file (Ctrl + S).
   • Note: Ensure you have a working telegram_message function defined elsewhere to send the report via Telegram.

3. Run the Program:

   • Ensure the virtual environment is activated (source venv/bin/activate).
   • In the Visual Studio Code terminal, run:

     python stock_report.py
     

   • Alternatively, use Visual Studio Code’s Run feature (Run > Run Without Debugging or Ctrl + F5).
   • The program will:
     • Fetch stock data for AAPL, MSFT, NVDA, AMZN, and TSLA.
     • Generate a report using the Mistral model.
     • Print the report in the terminal.
     • Save it to stock_report.txt in your project folder.
     • Send it via Telegram.

Step 4: Schedule the Script with Cron

1. Open the Crontab Editor:

   • In the Visual Studio Code terminal, open the crontab editor:

     crontab -e
     

2. Add a Cron Job:

   • Add the following line to schedule the script to run daily at 6:00 PM (replace /path/to/your/project with your actual project folder path):

     0 18 * * * /bin/bash -c 'cd /path/to/your/project && source venv/bin/activate && python stock_report.py'
     

   • To find your project path, run pwd in the terminal while in your project folder and use that path.
   • For example, if your project is in /Users/jakegrim/YouTube_Demo, use:

     0 18 * * * /bin/bash -c 'cd /Users/jakegrim/YouTube_Demo && source venv/bin/activate && python stock_report.py'
     

3. Save and Exit:

   • In the vi editor, press Esc, type :wq, and press Enter to save and exit.

4. Verify the Cron Job:

   • Check that the cron job was added:

     crontab -l
     

   • The script will now run daily at 6:00 PM, generating and sending the stock report via Telegram.

Conclusion

You’ve successfully:

• Installed Ollama and run the Mistral model locally on macOS.
• Created a Python program to generate a daily stock report using live data and the Mistral model.
• Saved the report to a file and sent it via Telegram.
• Scheduled the program to run automatically using cron.

Now, you’ll receive a stock report every day at 6:00 PM via Telegram. Have fun exploring local AI and building awesome projects!