Connecting MLflow Hosted Models and Label Studio for Auto Labelling
Hey HN, I'm Jinen. I’m a PhD student working on DL interpretability & Optimization Theory. Before beginning my PhD, I worked at DagsHub fine tuning vision models for domain specific deployments. I wanted to use ML models to help label the data, based on Label Studio's ML Backends. The goal was to use a model registered and tracked on MLflow. I found it tedious and involved a lot of boilerplate code to pipeline an MLFlow registered model into Label Studio's ML backend . Part of the challenges was setting up the web server, adapting the model outputs and reading through a lot of documentation…
What it does
In the maker’s words, at launch
Hey HN, I'm Jinen. I’m a PhD student working on DL interpretability & Optimization Theory. Before beginning my PhD, I worked at DagsHub fine tuning vision models for domain specific deployments. I wanted to use ML models to help label the data, based on Label Studio's ML Backends. The goal was to use a model registered and tracked on MLflow. I found it tedious and involved a lot of boilerplate code to pipeline an MLFlow registered model into Label Studio's ML backend . Part of the challenges was setting up the web server, adapting the model outputs and reading through a lot of documentation on all three tools (MLflow, Label Studio, and DagsHub). So I spent some streamlining the process. The project is finally merged so I wanted to share it with you! Since DagsHub integrates both MLflow & Label Studio, it sets up an end-to-end pipeline for active learning. Overview of functionality: - Connects MLFlow-registered models into Label Studio. - Allows inference and labeling for your models with a single function call change. - Includes Pre-configured models for common tasks across vision / audio / text domains. - Makes it easy to customize with user-defined hooks - Integrates cleanly with DagsHub, making it straightforward to set up an active learning pipeline I wanted to make auto-labeling easy for ML engineers without needing to learn web development stuff. The setup is simple: 1. Clone the repo and build the Docker container 2. Run the container or use the orchestrator 3. Use DagsHub’s Python client to connect your MLflow model to Label Studio Would love for you all to try it out and share your thoughts. If anyone's interested in making it work independently of DagsHub, PRs are welcome! Repo: https://github.com/DagsHub/ls-configurable-model Docs: https://dagshub.com/docs/use_cases/auto_labeling/ Cheers :)
Does the same job
all alternatives →
- IMI'm building a simple way to package, collaborate, and deploy ML models2024 · kitops.ml · ▲9
Heya HN, after spending +1 year building an ML-driven analytics product (that didn't pan out unfortunately), I've pivoted to solving a problem my team and I found while building the previous product … why the hell is it so hard to move a model from a Jupyter notebook, to a development server, then to a production pipeline!? To solve this my team and I started the open source KitOps project under the Apache 2 license. KitOps includes the Kit CLI that uses a Kitfile manifest to create ModelKits: 1. The kit CLI packages your model, datasets, code, and configuration into an OCI compliant…
More dev tools this month
the category →
Open-source GTM skills for technical founders
Dev tools · 29d ago · gtmcofounder.com
OpenTrailPaper is open-source bike computer firmware for the LilyGO T5S3 4.7" E-Paper PRO. It supports offline maps, GPX routes, FIT recording and Bluetooth sensors.
Dev tools · 2d ago · opentrailpaper.com
Launched alongside, October 2024
the whole month →
Softr for Notion▲925Turn Notion databases into portals & apps with no code
Dev tools · 2024 · softr.io
One inbox for all your work discussions
Work · 2024 · generalcollaboration.com