The goal of this project is to provide a simple demonstration of the capabilities of AstraDB as it related to its vector database and search functionality, Hybrid Search, langchain integration for agent memory, index Analyzers, Storage Attached Index (SAI) for vector search filtering.
These capabilities are used for a vectorization of time series and support tickets for root cause analsyis based on available research papers.
Below is the screenshot of the demo.
It is recommended to create a python virtual environment using your tool of choice before installing the dependencies. The steps below are based on the virtual environment tool venv.
python -m venv .venv
. ./.venv/bin/activate
pip install -r requirements.txt
This application is built using streamlit
Before running the application make sure you have created your own .env file. See the sample_.env for information on the parameters required.
streamlit run app.py
- Samuel Matioli - astra-agent-memory repo
- Radovan Zvoncek work on Time Series Forecasting with Astra