Blueprints for recommender system architectures: 10th anniversary edition
In this blog post I summarize 4 existing architectural blueprints, and present a new one that, in my opinion, encompasses all the previous ones.
Link
Machine Learning
Machine learning is a field of artificial intelligence (AI) that is concerned with learning from data. Machine learning has three components:
- Supervised learning: Fitting predictive models using data for which outcomes are available.
- Unsupervised learning: Transforming and partitioning data where outcomes are not available.
- Reinforcement learning: on-line learning in environments where not all events are observable. Reinforcement learning is frequently applied in robotics.
Posts on machine learning
In the following posts, machine learning is applied to solve problems using R.
End-to-End ML Pipelines with MLflow: Tracking, Projects & Serving
Definitive tutorial for advanced use of MLflow
Link
LMOps
LMOps is a research initiative on fundamental research and technology for building AI products w/ foundation models, especially on the general technology for enabling AI capabilities w/ LLMs and Generative AI models.
Link
ML Systems Industrialization and MLOps
Industrialization of Data Consumption by ML systems during both experimentation (historical batch) and production (batch & stream) — grow above & beyond toy-ML-with-csv and single-threaded-pickle-flasked-deployment.
[Link]{https://medium.com/@sunil_iitb/ml-systems-industrialization-and-mlops-b30106974454}
MLOps: Managing AI models at Scale
Model building is excellent, but if we can’t deploy these models then it becomes useless.
Link
MosaicML StreamingDataset: Fast, Accurate Streaming of Training Data from Cloud Storage
Loading your training data becomes an escalating challenge as datasets grow bigger in size and the number of nodes scales. We built StreamingDataset to make training on large datasets from cloud storage as fast, cheap, and scalable as possible. Specially designed for multi-node, distributed training, StreamingDataset maximizes correctness guarantees, performance, and ease of use.
Link
The 2023 MAD (Machine Learning, Artificial Intelligence & Data) Landscape
It has been less than 18 months since we published our last MAD landscape, and it has been full of drama.
Link
Use the Dust Platform to design and deploy Large Language Model apps.
Dust apps rely on model providers to interact with large language models. You can setup your first model provider by clicking on the Providers pane and setting up the OpenAI provider. You’ll need to create an account at OpenAI and retrieve your API key.
Link
Using ChatGPT to build a Kedro ML pipeline
Talk with ChatGPT to build feature-rich solutions with a Streamlit frontend
Link
A guide to making the best ML pipeline using AutoGluon 2022 new features
how to get the best python open source model without overfitting
Link
Homemade Machine Learning
Link
Awesome Production Machine Learning
This repository contains a curated list of awesome open source libraries that will help you deploy, monitor, version, scale and secure your production machine learning
Link
BEAM
A crash course on using Beam to run an entire ML workflow, from development to deployment.
Link
A machine learning library for unsupervised time series anomaly detection
A machine learning library for unsupervised time series anomaly detection.
Link
Aqueduct: Orchestrate & manage production ML
Aqueduct gives you a simple Python-native API to define machine learning pipelines, the ability to deploy those pipelines on your existing infrastructure (e.g., Spark, Kubernetes, Lambda), and visibility into the code, data, and metadata associated with your workflows. Aqueduct is fully open-source and runs securely in your cloud.
Link
