Arpitkr/fraud-detection-ui

Migrate Groq model from Llama 3.1 8B to GPT OSS 20B

Update README

Update README

Update README: switch backend from HF to Render

Update fraud UI to use Render API

Switch Groq model and add quota-safe fallback

Update README.md

Update README.md

Update README.md

Update README.md

Update README.md

Add clean notebook copy for GitHub preview

Clear notebook outputs for GitHub preview

Update README with Streamlit deployment and natural language feature

Fixed UI Space to use Streamlit SDK

Added Natural Language tab. LLM parses transaction to features

Update README: add Agent Pipeline, suggested action, similar cases

Update README: add Agent Pipeline, suggested action, similar cases

Restore Dockerfile for FastAPI API

Fix Dockerfile to run Streamlit UI

Copy requirements_ui.txt to root for Docker

Restore Dockerfile for FastAPI

Restore Dockerfile for Streamlit UI

Added agent pipeline UI: suggested action + similar cases

Added agent pipeline UI: suggested action + similar cases

Fixed app_port to 7860 for FastAPI

Fixed requirements.txt for FastAPI backend

Fixed Dockerfile to run FastAPI instead of Streamlit

Added agent pipeline: similar cases + suggested action

added missing dependencies

Restoring HF Space config

Hidimg HF frontmatter from GitHub

Add emoji to fix GitHub frontmatter rendering

Fix HF Space color metadata

Fix HF Space README config

Removed unnecessary comments

Updated Readme

Add RAG system with fraud knowledge base

Add HF Space config with app_port

Fix UI requirements and Streamlit startup

Fix UI Space to run Streamlit

Fix: lazy init Groq client inside function

Fix: point UI to live API endpoint

Fix: add groq to requirements

Fix: minimal requirements for API only

Fix: uvicorn pointing to main:app

Fix: valid colorTo value

Fix: add HF space config header to README

Initial commit - clean deploy