Getting started with GPT-2

It’s a bit tricky to see where to begin with OpenAI’s (in)famous GPT-2 model. This blog post is our first in a small series about NLP. We hope it helps!

Getting Python

Our preferred way of installing and managing Python, particularly for machine learning tasks, is to use the Anaconda Environment.

⚠️ Anaconda’s environments don’t quite work like virtualenv, or other Python environment systems that you might be familiar with. They don’t store things in the location you specify, they store things in the system-wide Anaconda directory (e.g. on macOS in “/Users/USER/anaconda3/envs/”). Just remember that once you activate them, all commands are inside the environment.

Anaconda bundles a package manager, an environment manager, and a variety of other tools that make using and managing Python environments easier.

➡️ Download the Anaconda installer for Windows or macOS

Once you’ve installed Anaconda, following these instructions to make an Anaconda Environment to use with GPT-2.

➡️ Create a new Anaconda Environment named GPT2 and running Python 3.x (the version of Python you need to be running to work with GPT-2 at the moment):

conda create -n GPT2 python=3

➡️ Activate the Conda environment:

conda activate GPT2

Getting and using GPT-2

➡️ Clone the GPT-2 repository to your computer:

git clone https://github.com/openai/gpt-2.git

➡️ While inside the activated Conda environment you’re using, install the requirements specified inside the GPT-2 repository:

pip install --upgrade -r requirements.txt

➡️ Use the model downloading script that comes with the GPT-2 repository to download a pre-trained model:

python3 download_model.py 774M

⚠️ You can specify which of the various GPT-2 models you want to download. Above, we’re requesting a download of the 774 million parameter model, which is about 3.1 gigabytes. You can also request the smaller 124M or 355M models, or the much larger 1.5B model.

You might need to wait a little while as the script downloads the model. You’ll see something like this:

Fetching model.ckpt.data-00000-of-00001: 44%|███████ | 1.36G/3.10G [02:16<03:44, 7.77Mit/s]

➡️ You’ll then need to open the file src/interactive_conditional_samples.py using your favourite programming editor, and update the model_name to the one that you downloaded:

def interact_model(
    model_name='774M',
    seed=None,
    nsamples=1,
    batch_size=1,
    length=None,
    temperature=1,
    top_k=0,
    top_p=1,
    models_dir='models',
):

➡️ You can then execute the Python script:

python3 src/interactive_conditional_samples.py

This will fire up the model, allowing you to enter some text. You’ll eventually (likely after some warnings that you can ignore about your CPU and the version of TensorFlow) see something like this:

Model prompt >>>

Enter some text and press return (we recommend only a sentence or two), then wait a bit and see what the model generates in response. This could take a while, depending on your computer’s capabilities.

We’ll be back with a follow-up article, exploring how you can actually use GPT-2 for something useful. Stay tuned!


For more AI content, check out our latest book, Practical Artificial Intelligence with Swift! It covers using Swift for AI in iOS applications, using Apple’s CreateML, CoreML, and Turi Create. If you like filling your brain with words, why not fill them with ours?

🧶 Yarn Spinner 1.0

The popular open source narrative game development framework, Yarn Spinner, which is maintained by Secret Lab and a fabulous community, has reached version 1.0. As part of our 1.0 release, we’ve debuted 5 exciting new features:

  1. Compiled Scripts — Yarn Spinner now compiles to a binary format.
  2. Automatic Compiling — In Unity, your Yarn scripts will automatically be compiled when they change.
  3. Line Tagging — You can automatically add unique tags to lines of dialogue, and generate a .csv file to send to translators with the click of a button.
  4. Code Extension — There’s a syntax highlighting extension, available from the marketplace, for Visual Studio Code.
  5. No more .yarn.txt — The file extension is now .yarn! It was time.

We want Yarn Spinner to be the best tool that it can be. As part of that, we’ve launched a Patreon page, and we’d love for you to help support its development!

We’ve got big plans, so please check out the website, follow us on Twitter, support the Patreon if you can, and join our Narrative Game Development Slack. And we’d really appreciate it if you shared the news!

🧠 First Steps with Swift for TensorFlow

We just finished presenting at the inaugural TensorFlow World conference, in Santa Clara, California. Mars, Tim, and Paris presented what might be the first 3-hour tutorial session on the brand new Swift for TensorFlow machine learning platform.

This post serves as both a follow-up to that session (which was recorded, and will be posted soon — we’ll update this post when that happens) and a standalone guide and tutorial to get started with Swift for TensorFlow.

We’ll be posting follow-up tutorials, which will get more advanced, over the coming weeks. (In the mean time, check out our new book on Practical Artificial Intelligence with Swift!)

Getting Swift for TensorFlow

There are two ways to get Swift for TensorFlow that we’d recommend right now. The first is to use Google’s Colaboratory (Colab), an online data science and experimentation platform, which means you use it via a browser and a Jupyter Notebooks-like environment.

The second is to install it locally, using Docker.

If you use Windows, we recommend using Google Colab, and if you use Linux or macOS, we recommend installing using the Docker image (it’s much easier than Docker’s reputation might suggest!)

Installing Swift for TensorFlow with Docker

➡️ First, make a folder on your local system in which to store your Swift Jupyter notebooks. For example, mine is located at /Users/parisba/S4TF/notebooks. You don’t need to put anything in there, just make sure you’ve created it.

➡️ Download and install Docker: https://hub.docker.com/

We’re not going to explain this process much, because once it’s done you don’t need to think about Docker or any of this process again. If you want to learn how Docker works, there are plenty of sources online.

➡️ Now, clone the following git repository:

git clone https://github.com/google/swift-jupyter.git

➡️ Then, change directory into the cloned repository, and execute the following command:

docker build -f docker/Dockerfile -t swift-jupyter .

➡️ Then, to launch the Docker container and Jupyter notebooks, execute the following command:

docker run -p 8888:8888 --cap-add SYS_PTRACE -v /path/to/books:/notebooks swift-jupyter

⚠️ Note that you will need to replace the /path/to/books in the above with the path to folder on your local system that you created earlier.

➡️ Open the URL that is displayed in your terminal, similar to the following:

Copy/paste this URL into your browser when you connect for the first time,
    to login with a token:
        http://0.0.0.0:8888/?token=6693795258c11e5f22280811ddebd714267e1e662d66068e

➡️ You should see something that looks like the following screenshot:

➡️ You’re ready to go!

Using Google Colaboratory

You don’t need to do much to use Google Colaboratory!

➡️ Make sure you have a Google Account, and then head to Google Colab’s blank Swift notebook.

➡️ That’s it! You’re done.

Training a Model

In this example, we assemble a multilayer peceptron network that can perform XOR.

It’s not very useful, but it showcases how you build up a model using layers, and how to execute training with that model. XOR was one of the first stumbling blocks of early work with artificial neural networks, which makes it a great example for the power of modern machine learning frameworks.

It’s simple enough that you know whether it’s correct… which is why we’re doing it!

➡️ Create a new notebook, and import the TensorFlow framework:

import TensorFlow

To represent our XOR neural network model, we need to create a struct, adhering to the Layer Protocol (which is part of Swift For TensorFlow’s API). Ours is called XORModel.

Inside the model, we want three layers:

  • an input layer, to take the input
  • a hidden layer
  • an output layer, to provide the output

All three layers should be a Dense layer (a densely-connected layer) that takes an inputSize and an outputSize.

The inputSize specifies that the input to the layer is of that many values. Likewise outputSize, for the out of the layer.

Each will have an activation using an activation function determines the output shape of each node in the layer. There are many available activations, but ReLU and Sigmoid are common.

For our three layers, we’ll use sigmoid.

We’ll also need to provide a definition of our @differentiable func, callAsFunction(). In this case, we want it to return the input sequenced through (passed through) the three layers.

Helpfully, the Differentiable protocol that comes with Swift for TensorFlow has a method, sequenced() that makes this trivial.

➡️ To do this, add the following code:

struct XORModel: Layer
{
  var inputLayer = Dense<Float>(inputSize: 2, outputSize: 2, activation: sigmoid)
  var hiddenLayer = Dense<Float>(inputSize: 2, outputSize: 2, activation: sigmoid)
  var outputLayer = Dense<Float>(inputSize: 2, outputSize: 1, activation: sigmoid)
  
  @differentiable func callAsFunction(_ input: Tensor<Float>) -> Tensor<Float>
  {
    return input.sequenced(through: inputLayer, hiddenLayer, outputLayer)
  }
}

➡️ Then we need to create an instance of our XORModel Struct, which we defined above. This will be our model:

var model = XORModel()

Next, we need an optimiser, in this case we’re going to use stochastic gradient descent (SGD) optimiser, which we can get from the Swift for TensorFlow library.

➡️ Our optimiser is, obviously, for the model instance we defined a moment ago, and wants a learning rate of about 0.02:

let optimiser = SGD(for: model, learningRate: 0.02)

➡️ Now we need an array of type Tensor to hold our training data ([0, 0], [0, 1], [1, 0], [1, 1]):

let trainingData: Tensor<Float> = [[0, 0], [0, 1], [1, 0], [1, 1]]

➡️ And we need to label the training data so that we know the correct outputs:

let trainingLabels: Tensor<Float> = [[0], [1], [1], [0]]

➡️ To train, we’ll need a hyperparameter for epochs:

let epochs = 100_000

Then we need a training loop. We train the model by iterating through our epochs, and each time update the gradient (the 𝛁 symbol, nabla, is often used to represent gradient). Our gradient is of type TangentVector, and represents a differentiable value’s derivatives.

Each epoch, we set the predicted value to be our training data, and the expected value to be our training data, and calculate the loss using meanSquaredError().

Every so often we also want to print out the epoch we’re in, and the current loss, so we can watch the traning. We also need to return loss.

Finally, we need to use our optimizer to update the differentiable variables, along the gradient.

➡️ To do this, add the following code:

for epoch in 0..<epochs
{
    let 𝛁model = model.gradient { model -> Tensor<Float> in

        let Å· = model(trainingData)

        let loss = meanSquaredError(predicted: Å·, expected: trainingLabels)

        if epoch % 5000 == 0
        {
          print("epoch: \(epoch) loss: \(loss)")
        }
        return loss
    }

    optimiser.update(&model, along: 𝛁model)
}

➡️ Run the notebook! You should see something resembling the following output:

epoch: 0 loss: 0.25470454
epoch: 5000 loss: 0.24981761
epoch: 10000 loss: 0.2496698
...snip...
epoch: 95000 loss: 0.16970003

➡️ Test your (incredibly useful) XOR model by adding a cell to your notebook with the following code:

print(round(model.inferring(from: [[0, 0], [0, 1], [1, 0], [1, 1]])))

➡️ The output should be as follows:

[[0.0],
 [1.0],
 [1.0],
 [0.0]]

➡️ Congratulations! You just trained a machine learning model that can, badly, perform XOR.

We’ll be posting more Swift for TensorFlow material in the coming weeks! 🚀


For more Swift AI content, check out our latest book, Practical Artificial Intelligence with Swift! It covers using Swift for AI in iOS applications, using Apple’s CreateML, CoreML, and Turi Create. If you like filling your brain with words, why not fill them with ours?

If you want to learn a little more about Swift for TensorFlow, we recommend this session from TensorFlow World as a great starting point:

Installing Unity ML-Agents

⚠️ At O’Reilly AI Conference San Jose, attending our tutorial? This is for you! Complete all the steps in this document to be ready for the tutorial.

Want to explore the Unity Machine Learning Agents Toolkit (“ML-Agents”)? Here’s the easiest way to get up and running on Windows or macOS.

Unity ML-Agents is a great way to explore machine learning, whether you’re interested in building AI for games, or simulating an environment to solve a broader ML problem, why not try Unity’s ML-Agents?

We’ll be posting a variety of guides and material covering various aspects of Unity’s ML-Agents, but we thought we’d start with an installation guide!


To use ML-Agents, you’ll need to install three things:

  1. Unity
  2. Python and ML-Agents (and associated environment and support)
  3. The ML-Agents Unity project

Unity

Installing Unity is the easiest bit. We recommend downloading and using the official Unity Hub to manage your installs of Unity.

➡️ Download the Unity Hub for Windows or macOS

The Unity Hub allows you to manage multiple installs of different versions of Unity, and lets you select which version of Unity you open and create projects with.

⚠️ We recommend installing Unity 2019.2.4f1 for the tutorial at O’Reilly AI Conference. If you install a different version, we might not be able to help you.

If you don’t want to use the Unity Hub, you can download different versions of Unity for your platform manually:

➡️ Download a specific version of Unity for Windows or macOS

We strongly recommend that you use the Unity Hub to manage your Unity installs, as it’s the easiest way to stick to a specific version of Windows, and manage your installs. It really makes things easier.

If you like using command line tools, you can also try the U3d tool to download and manage Unity install’s from the terminal.

Python and ML-Agents

Our preferred way of installing and managing Python, particularly for machine learning tasks, is to use the Anaconda Environment.

⚠️ Anaconda’s environments don’t quite work like virtualenv, or other Python environment systems that you might be familiar with. They don’t store things in the location you specify, they store things in the system-wide Anaconda directory (e.g. on macOS in “/Users/USER/anaconda3/envs/”). Just remember that once you activate them, all commands are inside the environment.

Anaconda bundles a package manager, an environment manager, and a variety of other tools that make using and managing Python environments easier.

➡️ Download the Anaconda installer for Windows or macOS

Once you’ve installed Anaconda, following these instructions to make an Anaconda Environment to use with Unity ML-Agents.

➡️ First, download 🔗 this yaml file, and execute the following command (pointing to the yaml file you just downloaded):

conda env create -f /path/to/unity_ml.yaml

➡️ Once the new Anaconda Environment (named UnityML) has been created, activate it using the following command in your terminal:

conda activate UnityML

The yaml file we provided specifies all the Python packages, from both Anaconda’s package manager, as well pip, the Python package manager, that you need to make an environment that will work with ML-Agents.

Doing it manually

You can also do this manually (instead of asking Anaconda to create an environment based on our environment file).

⚠️ You do not need to do this if you created the environment with the yaml file, as above. If you did that go straight to “Testing the environment”, below.

➡️ Create a new Anaconda Environment named UnityML and running Python 3.6 (the version of Python you need to be running to work with TensorFlow at the moment):

conda create -n UnityML python=3.6

➡️ Activate the Conda environment:

conda activate UnityML

➡️ Install TensorFlow 1.7.1 (the version of TensorFlow you need to be running to work with ML-Agents):

pip install tensorflow==1.7.1

➡️ Once TensorFlow is installed, installing the Unity ML-Agents:

pip install mlagents

Testing the environment

➡️ To check everything is installed properly, run the following command:

mlagents-learn --help

You should see something that looks like the following image. This shows that everything is installed properly.

If you’re coming to our conference tutorial, you’re now ready to go.

The ML-Agents Unity Project

The best way to start exploring ML-Agents is to use their provided Unity project. To get it, you’ll need a copy of the Unity ML-Agents repository.

⚠️ You do not need to do this bit if you’re coming to our tutorial at the O’Reilly AI Conference. We will provide a project on the day.

➡️ Clone the Unity ML-Agents repository to your system (see the note below if you’re coming to our tutorial!):

git clone https://github.com/Unity-Technologies/ml-agents.git

⚠️ If you’re coming to our O’Reilly AI Conference tutorial, we will provide a project on the day.

You should now have a directory called ml-agents. This directory contains the source code for ML-Agents, a whole of lot useful configuration files, as well starting point Unity projects for you to use.

➡️ You’re ready to go! If you’re coming to our tutorial, you’ll need a slightly different project which we’ll help you out with on the day!

We’ll have another article on getting started (now that you’ve got it installed) next week!


In San Jose? At O’Reilly AI Conference? Attend our tutorial!