How to Use Llama 3 with PandasAI and Ollama Locally

---

---

AI Nuggets

Setting Up the Environment

1. Create a virtual environment using conda:

   conda create -n genai  
   

2. Activate the virtual environment:

   conda activate genai  
   

Installing Required Tools

1. Create a requirements.txt file and list the following libraries:
   • pandas
   • pandasai
   • streamlit
2. Install the required libraries using pip:

   pip install -r requirements.txt  
   

Initializing Llama 3 with Ollama

1. Start the Ollama service in the terminal:

   ollama serve  
   

2. Install the Llama 3 model using Ollama:

   ollama pull llama3  
   

3. List installed models to confirm Llama 3 is loaded:

   ollama list  
   

Setting Up the App with Streamlit

1. Create a Python file named app.py.
2. Import the necessary classes from pandasai and streamlit:

   from pandasai.llm.local_llm import LocalLLM  
   import streamlit as st  

3. Initialize the Llama 3 model with Ollama:

   model = LocalLLM(api_base="http://localhost:11434/v1", model="llama3")  

4. Set the title of the app using Streamlit:

   st.title("Data analysis with PandasAI")  

5. Run the app with Streamlit:

   streamlit run app.py  
   

Building the App Interface

1. Create a file uploader widget for CSV files:

   uploaded_file = st.file_uploader("Upload a CSV file", type=['csv'])  

2. Check if a file is uploaded and display the first three rows:

   if uploaded_file is not None:  
       import pandas as pd  
       data = pd.read_csv(uploaded_file)  
       st.write(data.head(3))  

3. Convert the dataset into a SmartDataframe:

   from pandasai import SmartDataframe  
   df = SmartDataframe(data, config={"llm": model})  

4. Create a text area for user input:

   prompt = st.text_area("Enter your prompt:")  

5. Create a button to generate responses:

   if st.button("Generate"):  
       if prompt:  
           with st.spinner("Generating response..."):  
               st.write(df.chat(prompt))  

Making Inferences with the App

1. Write prompts to interact with the dataset, such as:
   • “How many rows and columns are in the dataset?"
   • "What is the average age?"
   • "How many people have more than 3 siblings?"
   • "How many people died and how many survived?"
   • "Return the percentage of passengers by gender"
   • "Draw a bar chart of the sex column"
   • "Plot a pie chart of the pclass column"
   • "Visualize the distribution of the fare column"
   • "Draw the histogram of the age column"
   • "Draw the histogram of the fare column separated by the sex column"
   • "Draw the heatmap of numerical variables”

Tips and Reminders

• Ensure the prompts are clear and specific to get accurate outputs.
• It may take a few attempts to find the right prompts for the desired output.

Conclusion

• The video covers the use of PandasAI with Llama-3 by building an app with Streamlit.
• Remember to subscribe, like the video, and leave a comment.

Additional Information

• The video description and additional URLs are not provided in the transcript. Please refer to the video on YouTube for any links or further instructions: http://youtube.com/watch?v=_dDaNgBDoHY