energy-usage-anomaly-detection-assistant

The energy usage anomaly detection assistant is an LLM based agentic assistant that uses tools to detect anomalies in smart meter data and provide human readable alerts to the customer.

The customer interacts with the assistant through a streamlit application. The application is deployed to Azure, and can be accessed here. Please reach out to Robert Mcleod for the password to access the application.

Local Setup

To run the application locally, follow the steps below:

1. Clone the repository

git clone https://github.com/robertmcleod2/energy-usage-anomaly-detection-assistant.git
cd energy-usage-anomaly-detection-assistant

2. create a virtual environment with python version 3.12. For Conda:

conda create -n energy-usage-anomaly-detection-assistant python=3.12
conda activate energy-usage-anomaly-detection-assistant

3. Install the required packages

pip install -r requirements_dev.txt

4. Set up your local environment variables. Create a .env file in the root directory of the project, following the template in the .env_template file. The streamlit password can be set to any value. The OPENAI_API_KEY can be obtained from OpenAI, or reach out to Robert Mcleod for the key.

5. Run the streamlit application

streamlit run src/app.py

6. View the application in your browser at http://localhost:8501. The password is the one you set in the .env file.