luigi
dvc
luigi | dvc | |
---|---|---|
14 | 109 | |
17,327 | 13,139 | |
0.5% | 0.8% | |
6.3 | 9.6 | |
9 days ago | 5 days ago | |
Python | Python | |
Apache License 2.0 | Apache License 2.0 |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
luigi
-
Ask HN: What is the correct way to deal with pipelines?
I agree there are many options in this space. Two others to consider:
- https://airflow.apache.org/
- https://github.com/spotify/luigi
There are also many Kubernetes based options out there. For the specific use case you specified, you might even consider a plain old Makefile and incrond if you expect these all to run on a single host and be triggered by a new file showing up in a directory…
-
In the context of Python what is a Bob Job?
Maybe if your use case is “smallish” and doesn’t require the whole studio suite you could check out apscheduler for doing python “tasks” on a schedule and luigi to build pipelines.
-
Lessons Learned from Running Apache Airflow at Scale
What are you trying to do? Distributed scheduler with a single instance? No database? Are you sure you don't just mean "a scheduler" ala Luigi? https://github.com/spotify/luigi
-
Apache Airflow. How to make the complex workflow as an easy job
It's good to know what Airflow is not the only one on the market. There are Dagster and Spotify Luigi and others. But they have different pros and cons, be sure that you did a good investigation on the market to choose the best suitable tool for your tasks.
-
DevOps Fundamentals for Deep Learning Engineers
MLOps is a HUGE area to explore, and not surprisingly, there are many startups showing up in this space. If you want to get it on the latest trends, then I would look at workflow orchestration frameworks such as Metaflow (started off at Netflix, is now spinning off into its own enterprise business, https://metaflow.org/), Kubeflow (used at Google, https://www.kubeflow.org/), Airflow (used at Airbnb, https://airflow.apache.org/), and Luigi (used at Spotify, https://github.com/spotify/luigi). Then you have the model serving itself, so there is Seldon (https://www.seldon.io/), Torchserve (https://pytorch.org/serve/), and TensorFlow Serving (https://www.tensorflow.org/tfx/guide/serving). You also have the actual export and transfer of DL models, and ONNX is the most popular here (https://onnx.ai/). Spark (https://spark.apache.org/) still holds up nicely after all these years, especially if you are doing batch predictions on massive amount of data. There is also the GitFlow way of doing things and Data Version Control (DVC, https://dvc.org/) is taken a pole position there.
-
Data pipelines with Luigi
At Wonderflow we're doing a lot of ML / NLP using Python and recently we are enjoying writing data pipelines using Spotify's Luigi.
- Noobie who is trying to use K8s needs confirmation to know if this is the way or he is overestimating Kubernetes.
-
Open Source ETL Project For Startups
💡【About Luigi】 https://github.com/spotify/luigi Luigi was built at Spotify since 2012, it's open source and mainly used for getting data insights by showing recommendations, toplists, A/B test analysis, external reports, internal dashboards, etc.
- Resources/tutorials to help me learn about ETL?
-
Using Terraform to make my many side-projects 'pick up and play'
So to sum that up, I went from having nothing for my side-project set up in AWS to having a Kubernetes cluster with the basic metrics and dashboard, a proper IAM-linked ServiceAccount support for a smooth IAM experience in K8s, and Luigi deployed so that I could then run a Luigi workflow using an ad-hoc run of a CronJob. That's quite remarkable to me. All that took hours to figure out and define when I first did it, over six months ago.
dvc
-
My Favorite DevTools to Build AI/ML Applications!
Collaboration and version control are crucial in AI/ML development projects due to the iterative nature of model development and the need for reproducibility. GitHub is the leading platform for source code management, allowing teams to collaborate on code, track issues, and manage project milestones. DVC (Data Version Control) complements Git by handling large data files, data sets, and machine learning models that Git can't manage effectively, enabling version control for the data and model files used in AI projects.
-
Why bad scientific code beats code following "best practices"
What you’re describing sounds like DVC (at a higher-ish—80%-solution level).
https://dvc.org/
See pachyderm too.
-
First 15 Open Source Advent projects
10. DVC by Iterative | Github | tutorial
-
Exploring Open-Source Alternatives to Landing AI for Robust MLOps
Platforms such as MLflow monitor the development stages of machine learning models. In parallel, Data Version Control (DVC) brings version control system-like functions to the realm of data sets and models.
- ML Experiments Management with Git
-
Git Version Controlled Datasets in S3
I was using DVC (https://dvc.org/) for some time to help solve this but it was getting hard to manage the storage connections and I would run into cache issues a lot, but this solves it using git-lfs itself.
- Ask HN: How do your ML teams version datasets and models?
-
Exploring MLOps Tools and Frameworks: Enhancing Machine Learning Operations
DVC (Data Version Control):
- Evaluate and Track Your LLM Experiments: Introducing TruLens for LLMs
-
[D] Is there a tool to keep track of my ML experiments?
I have been using DVC and MLflow since then DVC had only data tracking and MLflow only model tracking. I can say both are awesome now and maybe the only factor I would like to mention is that IMO, MLflow is a bit harder to learn while DVC is just a git practically.
What are some alternatives?
Airflow - Apache Airflow - A platform to programmatically author, schedule, and monitor workflows
MLflow - Open source platform for the machine learning lifecycle
Kedro - Kedro is a toolbox for production-ready data science. It uses software engineering best practices to help you create data engineering and data science pipelines that are reproducible, maintainable, and modular.
lakeFS - lakeFS - Data version control for your data lake | Git for data
Apache Spark - Apache Spark - A unified analytics engine for large-scale data processing
Activeloop Hub - Data Lake for Deep Learning. Build, manage, query, version, & visualize datasets. Stream data real-time to PyTorch/TensorFlow. https://activeloop.ai [Moved to: https://github.com/activeloopai/deeplake]
mrjob - Run MapReduce jobs on Hadoop or Amazon Web Services
delta - An open-source storage framework that enables building a Lakehouse architecture with compute engines including Spark, PrestoDB, Flink, Trino, and Hive and APIs
Dask - Parallel computing with task scheduling
ploomber - The fastest ⚡️ way to build data pipelines. Develop iteratively, deploy anywhere. ☁️
Pinball
aim - Aim 💫 — An easy-to-use & supercharged open-source experiment tracker.