Press "Enter" to skip to content

Category: Machine Learning

Running Azure ML On-Premises via Azure Arc

Published 2021-07-09 by Kevin Feasel

Tsuyoshi Matsuzaki takes us through running Azure Machine Learning via Azure Arc:

First of all, you must run Azure Arc enabled Kubernetes on-premise or on 3rd party cloud. For running Arc-enabled Machine Learning later, use machines with more than 4 CPUs, since Arc-enabled ML requires enough resources.

In this post, I assume that we run KIND (Kubernetes in Docker) cluster on on-premise Ubuntu server. (For test purpose, I have used Ubuntu 18.04 on a single virtual machine in Azure, Standard D3 v2, which has 4 CPUs and 14 GB memory.)

Click through to see how it’s done.

Comments closed

Optimizing BERT Models on Google Colab

Published 2021-07-08 by Kevin Feasel

Kevin Jacobs fine-tunes some NLP processes:

BERT is a language model and can thus be used for predicting the next word in a sentence. Furthermore, BERT can be used for automatic summarization, text classification and many more downstream tasks. Google Colab provides you with a cloud-based environment on which you can train your machine learning models on a GPU. The downside is that your data is uploaded to the Google cloud. Google Colab gives you the opportunity to finetune BERT.

Click through to see how.

Comments closed

Using Filter Based Feature Selection in Text Analytics

Published 2021-07-02 by Kevin Feasel

Dinesh Asanka takes us through a text analytics technique in Azure Machine Learning:

There are two parameters to be defined in the Feature Hashing control. Hashing bitsize will define the maximum number of vectors. 10 hashing bitsize means 1,024 vectors (2^10). 1,024 vectors are more than enough even for the large volume text files. Next, we need to choose N-grams which is 2 as 2 is the optimal number for N-grams for most situations. A detailed description of N-Grams is given in the link given in the reference section.

After the vectors are generated, we do not need other text columns. Apart from the vectors, we need only the dependent attribute or the category column in this example. Therefore, we can remove the unnecessary attributes by Select Columns in dataset control. However, this control will show 1,024 vectors even though it is not available in the previous step, Feature Hashing. Therefore, you need to choose only the available attributes in the Feature Hashing control at the Select Columns in dataset control. In the above example, only 93 vectors were generated.

Click through to learn more.

Comments closed

Monitoring Power Virtual Agent Chatbots

Published 2021-07-01 by Kevin Feasel

Devin Knight has a video for us:

Power Virtual Agents empowers subject matter experts to build intelligent conversational bots, using a guided, no-code graphical interface. In this video you will learn how to monitor how successful your chatbots are at answering your users questions. Using the monitoring capability you will uncover areas of your chatbot that can be improved.

If I were familiar enough with Latin, I’d try a play on “Quis custodiet ipsos custodes?” with this.

Comments closed

Declarative MLOps with Ludwig

Published 2021-06-21 by Kevin Feasel

Jacqueline Cardoso announces a new version of Ludwig:

Ludwig abstracts away the complexity of combining all these disparate systems together through its declarative approach to structuring machine learning pipelines. Instead of writing code for your model, training loop, preprocessing, postprocessing, evaluation, and hyperparameter optimization, you only need to declare the schema of your data as a simple YAML configuration:

I’ve long been a fan of declarative approaches to problem-solving, so I’m going to need to dig into this a bit.

Comments closed

Using Lobe for Training ML Models

Published 2021-06-02 by Kevin Feasel

Chris Webb reviews a free tool from Microsoft:

The most impressive thing about it is not what it does but how it does it: a lot of tools claim to make machine learning easy for non-technical users but Lobe really is easy to use. My AI/ML knowledge is very basic but I got up and running with it extremely quickly.

To test it out I downloaded lots of pictures of English churches and trained a model to detect whether the church had a tower or a spire. After I labelled the pictures appropriately:

Click through for Chris’s findings. Looks like the only thing it does today is image classification, but more functionality is forthcoming.

Comments closed

AI versus ML versus Deep Learning

Published 2021-05-14 by Kevin Feasel

Holger von Jouanne-Diedrich asks the expert:

This is our 101’st blog post here on Learning Machines and we have prepared something very special for you!

Oftentimes the different concepts of data science, namely artificial intelligence (AI), machine learning (ML), and deep learning (DL) are confused… so we asked the most advanced AI in the world, OpenAI GPT-3, to write a guest post for us to provide some clarification on their definitions and how they are related.

We are most delighted to present this very impressive (and only slightly redacted) essay to you – enjoy!

The machine has learned about itself. This is where I’m glad I only believe weak AI is possible…

Comments closed

Plotting XGBoost Trees with R

Published 2021-04-30 by Kevin Feasel

Andrew Treadway shows off a method to visualize the results of training an XGBoost model:

In this post, we’re going to cover how to plot XGBoost trees in R. XGBoost is a very popular machine learning algorithm, which is frequently used in Kaggle competitions and has many practical use cases.

Let’s start by loading the packages we’ll need. Note that plotting XGBoost trees requires the DiagrammeR package to be installed, so even if you have xgboost installed already, you’ll need to make sure you have DiagrammeR also.

Click through for the process. H/T R-Bloggers.

Comments closed

Predicting Insurance Prices with ML.NET

Published 2021-04-12 by Kevin Feasel

Chandra Kudumula shows off ML.NET:

There are three ways to begin with ML.NET

– API Model: You can start ML.NET through a Framework API and write code in C# or F#
– GUI Model: Use ML.NET Model builder in Visual Studio.
– CLI Model: For cross-platform development like Mac and Linux, use ML.NET CLI.

Let’s get started with API Model for predicting the insurance premium using ML.NET Framework.

I’m using Microsoft (MS) Visual Studio 2019 and creating a Console Application. Be sure that you have the latest version of VS and that .NET 5 SDK is installed.

Click through for the demo in Visual Studio using C#.

Comments closed