Data Science Blog

Overwatch Ranked Data

I started playing Overwatch towards the end of Season 2 and I thought it would be interesting to start collecting data from my own ranked games. The goal was to maintain a dataset of Overwatch ranked data that I could analyze to better understand how skill rating (SR) changes as a function of, for example, win/loss streaks and medals. I have made the dataset public in case anyone wants to run their own statistics or visualizations. However, I only have a few seasons worth of data as I took a break from the game after Season 4, and picked it up again…

Continue Reading

Colour image classification (CIFAR-10) using a CNN

As I mentioned in a previous post, a convolutional neural network (CNN) can be used to classify colour images in much the same way as grey scale classification. The way to achieve this is by utilizing the depth dimension of our input tensors and kernels. In this example I'll be using the CIFAR-10 dataset, which consists of 32x32 colour images belonging to 10 different classes. You can see a few examples of each class in the following image from the CIFAR-10 website: Although previously I've talked about the Lasagne and nolearn packages (here and here), extending those to colour images…

Continue Reading

Visualizing Convolutional Neural Networks using nolearn

We previously talked about Convolutional Neural Networks (CNN) and how use them to recognize handwritten digits using Lasagne. While we can manually extract kernel parameters to visualize weights and activation maps (as discussed in the previous post), the nolearn package offers an easy way to visualize different elements of CNNs. nolearn is a wrapper around Lasagne (which itself is a wrapper around Theano), and offers some nice visualization options such as plotting occlusion maps to help diagnose the performance of a CNN model. Additionally, nolearn offers a very high level API that makes model training even simpler than with Lasagne. In…

Continue Reading

Handwritten digit recognition with a CNN using Lasagne

Following my overview of Convolutional Neural Networks (CNN) in a previous post, now lets build a CNN model to 1) classify images of handwritten digits, and 2) see what is learned by this type of model. Handwritten digit recognition is the 'Hello World' example of the CNN world. I'll be using the MNIST database of handwritten digits, which you can find here. The MNIST database contains grey scale images of size 28x28 (pixels), each containing a handwritten number from 0-9 (inclusive). The goal: given a single image, how do we build a model that can accurately recognize the number that…

Continue Reading

Automatic Python documentation with Sphinx autodoc and ReadTheDocs

Generating Python documentation for packages/modules can be quite time consuming, but there's a way to generate it automatically from docstrings. This post is mostly a summary of the fantastic guide by Sam Nicholls found here, but with one important addition (see the section on mocking). We'll be using the following: Sphinx - Python package for generating documentation Sphinx autodoc - Sphinx extension to generate documentation from docstrings ReadTheDocs - build and host documentation online Before you start, make sure you've written docstrings for your modules/functions/methods. This is the most time consuming part, but you should be commenting and documenting your…

Continue Reading

Overview of Convolutional Neural Networks (CNN)

Regular feed-forward artificial neural networks (ANN), like the type featured below, allow us to learn higher order non-linear features, which typically results in improved prediction accuracy over smaller models like logistic regression. However, artificial neural networks have a number of problems that make them less ideal for certain types of problems. For example, imagine a case where we wanted to classify images of handwritten digits. An image is just a 2D array of pixel intensity values, so a small 28x28 pixel image has a total of 784 pixels. If we wanted to classify this using an ANN, we would flatten…

Continue Reading

Configure (base) Spacemacs

I love using Spacemacs. I like how extensible Emacs is, but prefer the modal editing nature of Vi(m), and Spacemacs provides the best of both worlds through the use of Evil. Tinkering with my editor configuration to get the perfect setup is fun, and there are a few ways this can be achieved: Use base Emacs with customizations Use the full Spacemacs distribution Customize the base Spacemacs distribution Using base Emacs (1) and adding our own packages and configurations is doable but it takes a lot of work. Full Spacemacs (2) offers everything the distribution has to offer, but I…

Continue Reading

XOR Logic Gate – Neural Networks (3/3)

(Part 3 of a series on logic gates) We have previously discussed OR logic gates and the importance of bias units in AND gates. Here, we will introduce the XOR gate and show why logistic regression can't model the non-linearity required for this particular problem. As always, the full code for these examples can be found in my GitHub repository here. XOR gates output True if either of the inputs are True, but not both. It acts like a more specific version of the OR gate: Input 1 Input 2 Output 0 0 0 0 1 1 1 0 1…

Continue Reading

AND Logic Gate – Importance of bias units (2/3)

(Part 2 of a series on logic gates) Previously, we talked about simple OR gates and now we'll continue that discussion with AND gates, and specifically the role of bias units. We often neglect to consider the role bias plays in our models. We know that we should include bias units, but why? Here, I'll walk through a short example using an AND gate to highlight the importance of the bias unit. Bias units allow us to offset the model in the same way that an intercept allows us to offset a regression line. Imagine a simple AND gate. It…

Continue Reading

OR Logic Gate using Theano (1/3)

(Part 1 of a series on logic gates) Theano is a powerful Python library that provides some useful tools for machine learning, such as GPU training and symbolic differentiation of the cost function during gradient descent. It can be a bit challenging to understand how Theano works, so before jumping into more complex non-linear models, we can get to grips with Theano by implementing something simple like an OR gate. An OR gate receives 2 inputs and will output true if either of the inputs are true. So, there are 3 cases where an OR gate will output a true…

Continue Reading
  • 1
  • 2