Press "Enter" to skip to content

Category: Machine Learning

Patterns for ML Models in Production

Published 2019-06-19 by Kevin Feasel

Jeff Fletcher shows four patterns for productionalizing Machine Learning models, as well as some things to take care of once you’re in production:

Operational Databases
This option is sometimes considered to be  real-time as the information is provided “as its needed,” but it is still a batch method. Using our telco example, a batch process can be run at night that will make a prediction for each customer, and an operational database is updated with the most recent prediction. The call center agent software can then fetch this prediction for the customer when they call in, and the agent can take action accordingly.

Read on for more.

Comments closed

Hyperparameter Tuning with MLflow

Published 2019-06-14 by Kevin Feasel

Joseph Bradley shows how you can perform hyperparameter tuning of an MLlib model with MLflow:

Apache Spark MLlib users often tune hyperparameters using MLlib’s built-in tools CrossValidator and TrainValidationSplit.  These use grid search to try out a user-specified set of hyperparameter values; see the Spark docs on tuning for more info.

Databricks Runtime 5.3 and 5.3 ML and above support automatic MLflow tracking for MLlib tuning in Python.

With this feature, PySpark CrossValidator and TrainValidationSplit will automatically log to MLflow, organizing runs in a hierarchy and logging hyperparameters and the evaluation metric.  For example, calling CrossValidator.fit() will log one parent run.  Under this run, CrossValidator will log one child run for each hyperparameter setting, and each of those child runs will include the hyperparameter setting and the evaluation metric.  Comparing these runs in the MLflow UI helps with visualizing the effect of tuning each hyperparameter.

Hyperparameter tuning is critical for some of the more complex algorithms like random forests, gradient boosting, and neural networks.

Comments closed

TensorFrames: Spark Plus TensorFlow

Published 2019-06-13 by Kevin Feasel

Adi Polak gives us an introduction to TensorFrames:

In all TensorFrames functionality, the DataFrame is sent together with the computations graph. The DataFrame represents the distributed data, meaning in every machine there is a chunk of the data that will go through the graph operations/ transformations. This will happen in every machine with the relevant data. Tungsten binary format is the actual binary in-memory data that goes through the transformation, first to Apache Spark Java object and from there it is sent to TensorFlow Jave API for graph calculations. This all happens in the Spark Worker process, the Spark worker process can spin many tasks which mean various calculation at the same time over the in-memory data.

An interesting bit of turnabout here is that the Scala API is the underdeveloped one; normally for Spark, the Python API is the Johnny-Come-Lately version.

Comments closed

Using the ML.NET Model Builder

Published 2019-06-12 by Kevin Feasel

I have a post looking at the ML.NET Model Builder:

You have four options from which to choose: two-class classification, multi-class classification, regression, or Choose Your Own Adventure. Today, we’re going to create a two-class classification model. Incidentally, they’re not kidding about things changing in preview—last time I looked at this, they didn’t have multi-class classifiers available.

Once you select Sentiment Analysis (that is, two-class classification of text), you can figure out how to feed data to this trainer.

I think this is fine for developers who are looking to add a machine learning component as a small part of a bigger product. I don’t think it will beat a trained human using R or Python, but it’s an interesting avenue.

Comments closed

Combining Machine Learning with DevOps

Published 2019-06-11 by Kevin Feasel

Rolf Tesmer explains that machine learning and DevOps aren’t oil and water (or maybe they are and we just need to stir harder):

In talking with various development teams, customers and DevOps engineers, a lot of the potential problems of meshing ML development into an enterprise DevOps process can be boiled down to a few different areas this aims to address…

– ML stack might be different from rest of the application stack
– Testing accuracy of ML model
– ML code is not always version controlled
– Hard to reproduce models (ie explainability)
– Need to re-write featurizing + scoring code into different languages
– Hard to track breaking changes
– Difficult to monitor models & determine when to retrain

So DevOps helps with this, right? Right?

Well er, some of them yes, but not all.

DevOps is not a panacea but it can solve certain types of problems well.

Comments closed

MLflow 1.0 Released

Published 2019-06-10 by Kevin Feasel

Clemens Mewald and Matei Zaharia announce the release of MLflow 1.0:

Today we are excited to announce the release of MLflow 1.0. Since its launch one year ago, MLflow has been deployed at thousands of organizations to manage their production machine learning workloads, and has become generally available on services like Managed MLflow on Databricks. The MLflow community has grown to over 100 contributors, and the MLflow PyPI package download rate has reached close to 600K times a month. The 1.0 release not only marks the maturity and stability of the APIs, but also adds a number of frequently requested features and improvements.

The release is publicly available starting today. Install MLflow 1.0 using PyPl, read our documentation to get started, and provide feedback on GitHub. Below we describe just a few of the new features in MLflow 1.0. Please refer to the release notes for a full list.

And it looks like they’re going to keep pushing on it from there.

Comments closed

Creating Models with ML.NET

Published 2019-06-05 by Kevin Feasel

I have a series on ML.NET; in this post, I look at building a model:

Okay, now that I have classes, I need to put in that lambda. I guess the lambda could change to qb => qb.Quarterback == "Josh Allen" ? "Josh Allen" : "Nate Barkerson" and that’d work except for one itsy-bitsy thing: if I do it the easy way, I can’t actually save and reload my model. Which makes it worthless for pretty much any real-world scenario.

So no easy lambda-based solution for us. Instead, we need a delegate. 

The experience so far has been a bit frustrating compared to doing similar work in R, but they’re actively working on the library, so I’m hopeful that there will be improvements. In the meantime, I’ve landed on the idea of doing all data cleanup work outside of ML.NET and just use the simplest transformations.

Comments closed

Time Series Modeling with Gluon

Published 2019-06-04 by Kevin Feasel

Jan Gasthaus, et al, announce a new open source product release:

We are excited to announce the open source release of Gluon Time Series (GluonTS), a Python toolkit developed by Amazon scientists for building, evaluating, and comparing deep learning–based time series models. GluonTS is based on the Gluon interface to Apache MXNet and provides components that make building time series models simple and efficient.

In this post, I describe the key functionality of the toolkit and demonstrate how to apply GluonTS to a time series forecasting problem.

It looks interesting.

Comments closed

Sentiment Analysis with Python

Published 2019-05-09 by Kevin Feasel

Bruno Stecanella shows us how to use MonkeyLearn to perform sentiment analysis in Python:

Sentiment analysis is a set of Natural Language Processing (NLP) techniques that takes a text (in more academic circles, a document) written in natural language and extracts the opinions present in the text.

In a more practical sense, our objective here is to take a text and produce a label (or labels) that summarizes the sentiment of this text, e.g. positive, neutral, and negative.

For example, if we were dealing with hotel reviews, we would want the sentence ‘The staff were lovely‘ to be labeled as Positive, and the sentence ‘The shared bathroom was absolutely disgusting‘ labeled as Negative.

Click through for a demo.

Comments closed

Scalable Anomaly Detection with Kafka and Cassandra

Published 2019-04-30 by Kevin Feasel

Paul Brebner wraps up a series on anomaly detection at scale:

The complete machine for the biggest result (48 Cassandra nodes) has 574 cores in total.  This is a lot of cores! Managing the provisioning and monitoring of this sized system by hand would be an enormous effort. With the combination of the Instaclustr managed Cassandra and Kafka clusters (automated provisioning and monitoring), and the Kubernetes (AWS EKS) managed cluster for the application deployment it was straightforward to spin up clusters on demand, run the application for a few hours, and delete the resources when finished for significant cost savings. Monitoring over 100 Pods running the application using the Prometheus Kubernetes operator worked smoothly and gave enhanced visibility into the application and the necessary access to the benchmark metrics for tuning and reporting of results.

The system (irrespective of size) was delivering an approximately constant 400 anomaly checks per second per core.

This is a good summary of what was an interesting series.

Comments closed