Enhanced Image Analysis with Ollama Vision Model

This Streamlit app allows users to analyze images using the Ollama Vision Model, with features such as conversation management.

Requirements

• Python 3.x
• Streamlit
• Ollama
• Pillow

Setup

1. Clone the repository:

   git clone https://github.com/agituts/ollama-vision-model-enhanced.git
   cd ollama-vision-model-enhanced

2. Set Up a Virtual Environmentt (Recommended):

   Create a virtual environment to isolate project dependencies.

   • On Windows

   python -m venv venv
   venv\Scripts\activate

   • On macOS/Linux

   python3 -m venv venv
   source venv/bin/activate

3. Install Python Dependencies:

   Install the required Python packages using pip:

   pip install -r requirements.txt

   If requirements.txt is missing or incomplete, install the packages manually:

   pip install streamlit ollama Pillow

4. Install and Set Up Ollama:

   • Download and Install Ollama: Visit https://ollama.com/ and follow the installation instructions for your operating system.

   • Verify Ollama Installation:

   ollama version

5. Pull the Required Ollama Model:

   • Pull the llama2:3b-vision model using Ollama:

   ollama pull llama2:3b-vision

6. Start the Ollama server:

   • Start the Ollama server to enable communication with the model:

   ollama serve

   Tips:

   • Keep this terminal window open as the server needs to run continuously. You may open a new terminal window for the next steps.
   • Also, in Windows, you will need to exit any instance of Ollama running in the taskbar.

7. Launch the App:

   • In a new terminal window (with your virtual environment activated), navigate to the project directory if you're not already there:

    cd ollama-vision-model-enhanced

   • Launch the app:

   streamlit run app.py

   This command will start the Streamlit server and open the app in your default web browser. If it doesn't open automatically, you can manually visit http://localhost:8501 in your browser.

Usage

Basic Operations:

• Upload an Image: Use the file uploader to select and upload an image (PNG, JPG, or JPEG).
• Add Context (Optional): In the sidebar under "Conversation Management", you can add any relevant context for the conversation.
• Enter Prompts: Use the chat input at the bottom of the app to ask questions or provide prompts related to the uploaded image.
• View Responses: The app will display the AI assistant's responses based on the image analysis and your prompts.

Conversation Management

• Save Conversations: Conversations are saved automatically and can be managed from the sidebar under "Previous Conversations".
• Load Conversations: Load previous conversations by clicking the folder icon (📂) next to the conversation title.
• Edit Titles: Edit conversation titles by clicking the pencil icon (✏️) and saving your changes.
• Delete Conversations: Delete individual conversations using the trash icon (🗑️) or delete all conversations using the "Delete All Conversations" button.

Troubleshooting

Issue: Ollama Model Not Found

Symptoms:

• Errors indicating the model cannot be found.
• The app fails to generate responses.

Solution:

• Ensure you've pulled the correct model with the exact name used in the code.
• Double-check the model name in the process_image_and_text function:

# Verify model name in code
response = ollama.chat(
    model='llama3.2-vision',   
)

Issue: Connection Error with Ollama Server

Symptoms:

• Errors related to connecting to the Ollama server.
• The app is unable to process image and text prompts.

Solution:

• Ensure the Ollama server is running in a terminal (ollama serve).
• Verify there are no firewall restrictions blocking communication.
• Restart the Ollama server if necessary.

Issue: Missing Python Packages

Symptoms:

• Import errors when running the app (e.g., ModuleNotFoundError).
• The app fails to start due to missing packages.

Solution:

• Ensure all dependencies are installed:

pip install -r requirements.txt

• If using a virtual environment, ensure it's activated when installing packages.

Issue: Streamlit App Not Starting

Symptoms:

• Terminal shows errors when running streamlit run app.py.
• The app doesn't open in the browser.

Solution:

• Verify that you're in the correct directory (ollama-vision-model-enhanced).
• Ensure app.py exists in the directory.
• Check for syntax errors or typos in app.py.

Additional Tips

Ollama Server:

• The Ollama server needs to run continuously while you're using the app.
• If you close the terminal or the server stops, restart it with:

ollama serve

Running on a Different Port:

• If you need to run the Streamlit app on a different port:

streamlit run app.py --server.port <PORT_NUMBER>

Stopping the App:

• To stop the Streamlit app, press Ctrl+C in the terminal where it's running. Updating the App Code:
• If you make changes to app.py, Streamlit will prompt you to rerun the app. Click "Rerun" or press R in the terminal.