Lazy Loading Integrations in ZenML
November 26, 2021 - Baris Can Durak
When working on a project in Python, it is very likely that you will run into an issue where even the simplest of imports can lead to a chain of imports, which in turn can cost you a few seconds of run time before you even start to use what you imported. Left unchecked, this eager consumption of time can become even more apparent and annoying if you working on a project where the response time is critical and there are a wide variety of tools in play. Let’s put this into perspective for a tool that handles Machine Learning workflows in a production setting.
Pipeline Conversations: Our New Podcast
November 19, 2021 - Adam Probst
This first episode of Pipeline Conversations is the kick-off for a series of talks in the broader Machine Learning Start-Up Space. Our goal is not to promote any product or company but to discuss and also educate the community on different topics. We will invite thought leaders in the MLOps space, talk about how to build an open-source startup in public, and also the biggest challenges in Machine Learning. There won’t be a strict script and our guests will mainly determine the content.
Why ML should be written as pipelines from the get-go
March 31, 2021 - Hamza Tahir
Today, Machine Learning powers the top 1% of the most valuable organizations in the world (FB, ALPH, AMZ, N etc). However, 99% of enterprises struggle to productionalize ML, even with the possession of hyper-specific datasets and exceptional data science departments.
Why ML in production is (still) broken - [#MLOps2020]
June 26th, 2020 - Hamza Tahir
Just a few days ago, I was able to share my thoughts on the state of Machine Learning in production, and why it’s (still) broken, on the MLOps World 2020. Read on for a writeup of my presentation, or checkout the recording of the talk on Youtube.
Can you do the splits?
June 11th, 2020 - Hamza Tahir
One attempt to ensure that ML models generalize in unknown settings is splitting data. This can be done in many ways,
from 3-way (train, test, eval)
splits to k-splits with cross-validation. The underlying reasoning is that by training a ML model
on a subset of the data, and evaluating on unknown data, one can reason much better if the model has underfit or overfit in training.
A case for declarative configurations for ML training
May 17th, 2020 - ZenML Team
No way around it: I am what you call an “Ops guy”. In my career I admin’ed more servers than I’ve written code. Over twelve years in the industry have left their permanent mark on me. For the last two of those I’m exposed to a new beast - Machine Learning. My hustle is bringing Ops-Knowledge to ML. These are my thoughts on that.
Why deep learning development in production is (still) broken
May 1st, 2020 - Hamza Tahir
Around 87% of machine learning projects do not survive to make it to production. There is a disconnect between machine learning being done in Jupyter notebooks on local machines and actually being served to end-users to provide some actual value.
Distributed PCA using TFX
February 27, 2020 - Hamza Tahir - Crossposted on Tensorflow Blog
Principal Component Analysis (PCA) is a dimensionality reduction technique, useful in many different machine learning scenarios. In essence, PCA reduces the dimension of input vectors in a way that retains the maximal variance in your dataset. Reducing the dimensionality of the model input can increase the performance of the model, reduce the size and resources required for training, and decrease non-random noise.