.Net Foundation ML.NET, 23 Sep 2019

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Getting started with ML.NET | ML.NET [3 of 8]

Description

Learn how you can get started with ML.NET on Windows, Mac, and Linux using tools (ML.NET Model Builder, ML.NET CLI) or code-first using ML.NET API.

Follow Pranav Rastogi: https://twitter.com/rustd

Watch the entire series here: https://aka.ms/dotnet3-mlnet
Download Visual Studio 2019 here: https://aka.ms/VisualStudio-dotnetconf19
Download .NET Core 3.0 here: https://aka.ms/Dotnet3

Learn more about ML .NET:
Get started with ML.NET: https://dotnet.microsoft.com/learn/ml-dotnet/get-started-tutorial/intro?&WT.mc_id=Educationalmlnet-c9-niner
What is Machine Learning with .NET? https://dotnet.microsoft.com/apps/machinelearning-ai/ml-dotnet?&WT.mc_id=Educationalmlnet-c9-niner
ML.NET Model Builder: https://dotnet.microsoft.com/apps/machinelearning-ai/ml-dotnet/model-builder?&WT.mc_id=Educationalmlnet-c9-niner
GitHub Repo ML .NET: https://github.com/dotnet/machinelearning
GitHub Repo ML.NET Samples: https://github.com/dotnet/machinelearning-samples/blob/master/README.md

Follow us:
 Twitter: https://twitter.com/dotnet
Facebook: https://www.facebook.com/Dotnet/
Blogs: https://devblogs.microsoft.com/dotnet/?WT.mc_id=Educationalmlnet-c9-niner
Shows and Podcasts: https://channel9.msdn.com/Shows/On-NET/?&WT.mc_id=Educationalmlnet-c9-niner

Useful Links:
Learn .NET: https://dotnet.microsoft.com/learn?&WT.mc_id=Educationalmlnet-c9-niner
.NET Community: https://dotnet.microsoft.com/platform/community?&WT.mc_id=Educationalmlnet-c9-niner
.NET Documentation: https://docs.microsoft.com/en-us/dotnet/?&WT.mc_id=Educationalmlnet-c9-niner

A

In this video, we'll be looking at how to get started with ml.net.

A

There are three ways to use: ml.net.

A

You can use ml.net through a framework api.

A

So you can write code in c, sharp and f sharp.

A

So this is a quick first way of getting started..

A

You can get started with a visual tool called model builder, where inside of visual studio, you can use gui-based tool to build custom, machine learning, model.

A

You can also use the ml.net cli for doing.

A

Cross-Platform development on mac and linux, as well.

A

Let's look at what are the basics of machine learning, workflow.

A

There are three key tenants of the machine learning workflow.

A

One is around preparing your data..

A

The second is around building and training your machine learning model,, where you'll get to evaluate different algorithms with different settings and see, which is the best model.. The last part is going to be around,.

A

Once you have your model,, how do you consume the model in your application,, which could either be a web app,, a desktop app,, a mobile app, gaming, app or any kind of application.?

A

So there are four to five key steps.

A

The first two are around connecting to your data,, whether it's stored in a database or in a csv file and you're. Extracting your features, which is more relevant to your problem.

A

So for example, for sentiment analysis, you want to extract sentiment text as the features of your model.

A

Then you want to train the model on using different algorithms,.

A

Different settings, and then you want to evaluate the model which performs best for your scenario.

A

Lastly, you want to consume the model where you.

A

Can take this model and use it in any application.

A

Now we'll look at how can you get started with the machine learning workflow in ml.net across all the three ways that we have.

A

The first way that we look at is getting.

A

Started with ml.net using model builder.

A

So this is visual studio, 2019.

A

You can also use it in 2017 as well., where, once you install the model builder extension,.

A

A machine learning is enabled for all.net projects.

A

In this case, I have a console app., so I can right-click console app and say add machine learning,.

A

Which will launch model builder as.

A

A tabbed window in the visual studio, too.

A

I can start by choosing a scenario.

A

in this case,. The scenarios are around whether I want to do sentiment analysis,, where I want to classify sentiments as positive or negative.

A

I can do issue classification,, where I'm going to classify issues into three different categories or price prediction where I want.

A

To predict a price of a taxi fare or I can build my custom scenario as well.

A

So I'll choose sentiment: analysis.

A

The next step in the processes around data,.

A

Where you connect to your data,, you load your data and you extract the features which are relevant to your scenario..

A

In this case,. I can either connect to file or sql server.

A

So I'm going to choose file and I can upload either a csv or a tsv file.

A

So I can launch this., I can go to my desktop and I can upload the wikipedia dataset.

A

Now,, what this screen shows you is a sample of the first 10 rows of your data., so you can see what sentiment you get for that particular sentiment. Text..

A

The next step in the data process is choosing. What do you want to predict.

A

So in this case, I'm going to predict the sentiments..

A

This is the output of your machine learning model process.

A

So I'm going to choose sentiment as the column. I want to predict.

A

The third step is around training., so here, model builder uses automated machine learning under the covers, which explores various models and settings so that you, as a developer, don't have to worry.

A

About the intrinsics of machine learning, algorithms and setting and figuring out, which is the best model to use.

A

You can run this tool for a certain number of seconds.

A

The longer that you run it for, the better model that you'll get. the next step. You can evaluate the performance of this model.

A

So, in step, four,, what I'm showing over here is what was the current best model, which was sdca logistic regression binary..

A

This was the accuracy that I got and the tool also explored different models.

A

So here's a summary of all the different models that were tried.

A

And I can see what was the accuracy of the other models as well.

A

The last step in this process is around consuming the model.

A

So in this case, I can hit add projects,.

A

Which is going to add two projects to my solution..

A

One is a console app which has code for how the model was.

A

Trained in this model, builder cs and a class library,.

A

Which has the model large, zip file,, which is the machine learning model that was trained, that I can include in my application..

A

So I'm going to add a reference to the generated class library by saying "add reference".

A

Then,, I'm going to use this code from this screen.

A

In my console app where I'm going to try predicting sentiments.

A

So let's include a few namespaces here.

A

Let's include the namespace to the model, input, model output.

A

So in this case, I can say input.sentimenttext equals,. This is rude.

A

So here is where I can try out different kinds of input.

A

In my output, I can say console.writeline and I can predict.

A

The result of calling this model on this particular input.

A

So let's go ahead and run this application.

A

You'll see that the predicted sentiment.

A

Was true for this particular comment.

A

So we saw this was a very quick overview of how to get started with ml.net model builder as a way to build custom machine learning, model.

A

Our next step in this process is around getting started with the ml.net cli.

A

So we'll take a look at how you get started with ml.net on a mac or linux.

A

What I have over here, is the terminal and I've installed.

A

Ml.Net cli, which is a.net global tool.,.

A

So if I do ml net,, since this is the cli command- and I can say.

A

Ml net auto train, which will.

A

List out all the options that are possible on training, a machine learning model..

A

So let's go ahead and train custom machine learning model for sentiment, analysis.

A

What I have done over here, is, I've downloaded the wikipedia dataset, which is a set of sentiment and sentimenttext..

A

I'm going to go ahead and create a.net application as the first step.

A

So here, I'm going to.

A

Create a new app called the consume model console app.

A

So this is a.net console app. Where we'll add this machine learning model to our application.

A

Next step is around training the machine learning model.

A

So in this case, I'm going to use a command called ml net, auto train with.

A

The binary classification task and with a dataset as the wikipedia dataset.

A

in this case,, the label column name,, which is the column that you want to predict, or the output of your machine learning model as sentiment, and I'm going to train the model for 10 seconds.

A

now,. The longer that you train for,, the better model that you get..

A

When I start training what happens behind the covers. Is,.

A

The cli uses automated machine learning, which explores.

A

Different model with various settings to give you the best possible model for your scenario., so you as a developer, don't have to worry about learning the.

A

Intrinsic and fine tuning the machine learning models.

A

So in this case, I got a summary saying that they were about five models: tried.

A

The best models was average perceptron with about.

A

A 70 percent accuracy and it created two projects.

A

One,, a class library project to consume the model and second,.

A

A console app which has code for how the model was trained.

A

So let's go ahead and add this model library to our application.

A

So I'm going to go to my app,.

A

Which is the consumer model app and I'm going to do two steps? Here.

A

first is adding a reference to the class library and second, is copying. The model file to this application.

A

Let's go ahead and add.

A

A reference to the generative class library,.

A

And let's go ahead and add the machine learning model as well.

A

So we're going to copy the machine learning model and we're.

A

Going to add a reference to ml.net as a framework by using this command.

A

So now we're going to go back to.

A

Vs code and we're going to open this project that we just created.

A

So far, this project has nothing,.

A

So you can see just a console app and now we'll add.

A

Code to consume the model that we just added.

A

So this is the code to consume the model in a console, app.

A

I'll create a remote context,, which is a machine learning contexts for ml.net to do different, tasks.

A

First task is around loading, the model,, creating the prediction engine and then creating input. File,.

A

My model input where I'm going to give the sentiment.

A

That I want to predict and I'm going to output the sentiment of my application..

A

So now that we've added code to consume the model,.

A

Let's see what the output of this code looks like., so I'm going to go back to my terminal and I'm going to run this application..

A

So when I say.net run, it's going to take that text and it's going to predict whether.

A

This comment was toxic sentiment or not.

A

In this case,, it turns out that this is a toxic sentiment.

A

So we just saw how to get started with.

A

Ml.Net on a mac or a linux, machine.

A

Now that we're back in windows, we'll look at the third way of getting started with ml.net, which is through the ml.net api.

A

So, let's switch over to visual studio..

A

This is a console application which is.

A

A document core console app and I'm going to show you.

A

How to do the same scenario of sentiment, analysis using the ml.net api.

A

So this is the dataset that I have, which is the wikipedia dataset.

A

It has two columns sentiment and sentimenttext, and I'm going to train a machine learning model which is going to take all these sentimenttext and is going to predict what is the actual sentiment..

A

The first thing I want to show you as part of the machine learning workflow, is around loading data and extracting features.

A

As I do that,, we start by creating a new ml context.

A

Since this is a context that is going to hold all the ml.net intrinsics,, this is similar to entity framework contexts or any other contexts that you're used to in the.net ecosystem.

A

Then I'm going to load my data into the ml context, pipeline.

A

The first step over here is around loading. This wikipedia dataset.

A

I have defined my model input, type,.

A

Which is going to be, what does the input columns look like.

A

In this case, I have two columns sentiment and sentimenttext.

A

So I'm going to load the data into these types.

A

The second step around in the machine learning process after you've loaded your data,.

A

Is around extracting features and choosing a machine learning algorithm., so in this case, the only features I have is the sentimenttext,.

A

So I'm going to do text featurization..

A

The next step is around adding an algorithm for binary classification.

A

In this case, I'm going to choose a trainer called sdc and logistic regression,.

A

Specify what is the output column? I want to predict, in this case,.

A

Sentiment and give it the input, features for training.

A

So in this case, sentimenttext.

A

I'm going to train this particular machine.

A

Learning model and then I'm going to evaluate the performance of this model by checking what was the accuracy that I get of this model.

A

The last step in this process is around trying to model.

A

So once I've trained the model,, I can create prediction engine which is.

A

A prediction engine I can use to predict the outcome of a certain thing.

A

In this case, I'm going to create.

A

A sample input that I'm going to call in the prediction: engine,.

A

Which is going to return me?

A

What was the result of this prediction.?

A

So let's go ahead and run this particular application.

A

So the steps that this is going to go through is,, it's going to load the data,, it's going to train the model and then it's going to create.

A

The prediction engine pool and it's going to try the model with a sample input that I gave.

A

So in this case,, what it.

A

Predicted for this input text's was ruled,.

A

Was this a toxic sentiment.

A

So just to summarize, this was an easy way to get started with the ml.net.

A

Api, where the first step is around creating the ml contexts.

A

The second step is around loading data,.

A

And then around extracting features and adding a machine learning algorithm.

A

Next step is around training the machine learning model and then evaluating the machine learning model..

A

The last step is around trying the machine learning model with your custom, data.

A

So in this series of video we saw how to get started with ml.net.

A

There are three ways that we saw, through the code-first approach, using the ml.net api,.

A

A visual studio to ui based approach using.

A

Model builder and a cross-platform approach using the ml.net cli.

A

In the next video, we will look at how to build a custom machine learning model for doing sentiment. Analysis.