power-to-the-brand/clustering-service

feat: handle scenario to convert pdf to markdown given a url

feat: add page range feature for convert pdf & remove skip_first_page feature

feat: add pdf markdown conversion api using pymupdf4llm

feat: remove pyspark

feat: add python-multipart

fix: add default prompt name

feat: add clustering csv

feat: update record token usage

fix: description in merge cluster

fix: bug for merging cluster

chore: revert to openai embedding model

feat: add voyageai embedding model & merge clusters api

feat: update prompt

feat: update summarize prompt

chore: missing import

feat: update cluster feedback request

feat: update api for process feedback

feat: add clusters_feedback API

feat: use strict mode and update model to gpt-4.1-mini

feat: add evaluate doc relationship API

feat: add more details

feat: add deepeval eval summarization APIs

feat: handle to cluster using embeddings in json file

feat: get prompt from langfuse

feat: add asyncio

fix: return 202

fix: should use PUT

fix: revert back to openai embedding model

feat: added the callback_url setting for clustering service

feat: add file input api and switch to google gemini

feat: update code for adaptive resolution

feat: add new approach

fix: update summarize prompt

feat: update prompt

feat: update prompt

feat: add logs for summarizing

feat: update using Faiss

feat: revert

feat: experiment create faster graph approach

feat: add sample dataset endpoint

feat: added readme.md

feat: initial commit for clustering service using igraph