BLOG
Get inspired by our latest blog articles.
Serving ML models at scale using Mlflow on Kubernetes – Part 3
26 October 2021
This article is the third part of a series in which we go through the process of logging models using Mlflow, serving them on Kubernetes engine and finally scaling them up according to our application needs. Although this article could be used independently to test any API response, we recommend reading our two previous articles (part1 and part2) on how to deploy a tracking instance and serve a model as an API with Mlflow. In the following, we will be interested in the scalability issue and address it with few experiments to understand k8s cluster behavior and give recommendations on how to handle high loads.
This article is the third part of a series in which we go through the process of logging models using Mlflow, serving them on Kubernetes engine and finally scaling them up according to our application needs. Although this article could be used independently to test any API response, we recommend reading our two previous articles (part1 and part2) on how to deploy a tracking instance and serve a model as an API with Mlflow. In the following, we will be interested in the scalability issue and address it with few experiments to understand k8s cluster behavior and give recommendations on how to handle high loads.
Serving ML models at scale using Mlflow on Kubernetes – Part 2
25 October 2021
This article is the second part of a series in which we go through the process of logging models using Mlflow, serving them as an API endpoint, and finally scaling them up according to our application needs. We encourage you to read our previous article in which we show how to deploy a tracking instance on k8s and check the hands-on prerequisites (secrets, environment variables) as we will continue to build upon them here.
In the following, we show how to serve a machine learning model that is already registered in Mlflow and expose it as an API endpoint on k8s.
This article is the second part of a series in which we go through the process of logging models using Mlflow, serving them as an API endpoint, and finally scaling them up according to our application needs. We encourage you to read our previous article in which we show how to deploy a tracking instance on k8s and check the hands-on prerequisites (secrets, environment variables) as we will continue to build upon them here.
In the following, we show how to serve a machine learning model that is already registered in Mlflow and expose it as an API endpoint on k8s.
Serving ML models at scale using Mlflow on Kubernetes – Part 1
22 October 2021
MLflow is a commonly used tool for machine learning experiments tracking, models versioning, and serving. In our first article of the series “Serving ML models at scale”, we explain how to deploy the tracking instance on Kubernetes and use it to log experiments and store models.
MLflow is a commonly used tool for machine learning experiments tracking, models versioning, and serving. In our first article of the series “Serving ML models at scale”, we explain how to deploy the tracking instance on Kubernetes and use it to log experiments and store models.
Visual time series forecasting with Streamlit Prophet
22 september 2021
You need a baseline for your latest time series forecasting project? You want to explain the decision-making process of a predictive model to a business audience? You would like to understand if car prices are seasonal before buying a new one? We might have something for you! This article introduces Streamlit Prophet, a web app to help data scientists train, evaluate and optimize forecasting models in a visual way. Forecasts are made with Prophet, a fast and easily interpretable model.
You need a baseline for your latest time series forecasting project? You want to explain the decision-making process of a predictive model to a business audience? You would like to understand if car prices are seasonal before buying a new one? We might have something for you! This article introduces Streamlit Prophet, a web app to help data scientists train, evaluate and optimize forecasting models in a visual way. Forecasts are made with Prophet, a fast and easily interpretable model.
The path to developing a high-performance demand forecasting model — Part 3
10 september 2021
How to choose the right visualizations and implement them in Streamlit to better debug your forecasting models
How to choose the right visualizations and implement them in Streamlit to better debug your forecasting models
Demand forecasting: Using machine learning to predict retail sales
24 August 2021
All industries aim to manufacture just the right number of products at the right time, but for retailers this issue is particularly critical as they also need to manage perishable inventory efficiently
All industries aim to manufacture just the right number of products at the right time, but for retailers this issue is particularly critical as they also need to manage perishable inventory efficiently
The $2 Trillion Club: how Data x Services are leading to skyrocketing growth in company valuations
5 August 2021
Recently, Microsoft joined Apple as the second listed US company (and third in the world, alongside Saudi oil giant Aramco) as a member of the very select $2 trillion club. The fact that US tech giants are dominating the world is not news, but the pace at which they’re crossing significant valuation milestones is. It took less than 18 months for them to move up from $1 trillion to $2 trillion. How is this even possible? What can “legacy” companies take away from this unparalleled economic success?
Recently, Microsoft joined Apple as the second listed US company (and third in the world, alongside Saudi oil giant Aramco) as a member of the very select $2 trillion club. The fact that US tech giants are dominating the world is not news, but the pace at which they’re crossing significant valuation milestones is. It took less than 18 months for them to move up from $1 trillion to $2 trillion. How is this even possible? What can “legacy” companies take away from this unparalleled economic success?
AI Requires a Holistic Framework and Scalable Projects
28 July 2021
Artificial intelligence and digital transformation projects have a low success rate, but best practices help.
Artificial intelligence and digital transformation projects have a low success rate, but best practices help.
Including ethics best practices in your Data Science project from day one
27 July 2021
Here are some guidelines to build trustworthy machine learning solutions without falling into ethical pitfalls.
Here are some guidelines to build trustworthy machine learning solutions without falling into ethical pitfalls.
Load More