Skip to content

Commit

Permalink
Create README.md
Browse files Browse the repository at this point in the history
  • Loading branch information
mjeremyhsu authored Aug 14, 2024
1 parent 3563fd4 commit 9e9373f
Showing 1 changed file with 23 additions and 0 deletions.
23 changes: 23 additions & 0 deletions README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,23 @@
# mKB - mimik Knowledge Base Microservice
mKB is a microservice designed for performing semantic searches on datasets. These datasets are pre-processed, chunked, and indexed by creating vector representations (embeddings) of the documents using the mimik dataset-cli tool.

## Features

### Dataset Management

- Create Datasets: Upload pre-processed JSON datasets to the Knowledge Base.
- List Datasets: Retrieve a list of all available datasets.
- Delete Datasets: Remove datasets by their name.
- Upload Chunks: Add specific chunks to an existing dataset using a pre-processed file.

### Search and Retrieval

- Similarity Search: Perform searches across datasets based on a prompt, returning the most relevant content.
- Chunk Retrieval: Retrieve individual chunks by their chunkId for detailed inspection.

## Installation Guide

1. Download the latest .zip file for the most recent release [HERE](https://github.com/edgeMicroservice/mKB/releases).
2. Unzip the package.
3. Locate the unzipped .tar file; this is the artifact needed for the edge microservice deployment.
4. If you are deploying the edge microservice for the first time, please follow the [edge microservice deployment quick start guide](https://devdocs.mimik.com/tutorials/01-index).

0 comments on commit 9e9373f

Please sign in to comment.