Data Science

As a data scientist, you deal with a lot of data. For small datasets, maybe you just store this information in a CSV file and load it into Pandas [https://pandas.pydata.org/]. However, this isn't really a scalable solution and won&

In my introductory post [https://www.jeremyjordan.me/neural-networks-representation/] on neural networks, I introduced the concept of a neural network that looked something like this. As it turns out, there are many different neural network architectures [http://www.asimovinstitute.org/neural-network-zoo/], each with its

In previous posts, I've introduced the concept of neural networks [https://www.jeremyjordan.me/neural-networks-representation/] and discussed how we can train neural networks [https://www.jeremyjordan.me/neural-networks-training/]. For these posts, we examined neural networks that looked like this. However, many of

So you've built a machine learning model and trained it on some data... now what? In this post, I'll discuss how to evaluate your model, and practical advice for improving the model based on what we learn evaluating it. I&

In my first post on neural networks, I discussed a model representation for neural networks [https://www.jeremyjordan.me/neural-networks-representation/] and how we can feed in inputs and calculate an output. We calculated this output, layer by layer, by combining the inputs from the

Gradient descent is an optimization technique commonly used in training machine learning algorithms. Often when we're building a machine learning model, we'll develop a cost function which is capable of measuring how well our model is doing. This function will

Principal components analysis (PCA) is the most popular dimensionality reduction technique to date. It allows us to take an $n$-dimensional feature-space and reduce it to a $k$-dimensional feature-space while maintaining as much information from the original dataset as possible in the reduced

Activation functions are used to determine the firing of neurons in a neural network. Given a linear combination of inputs and weights from the previous layer, the activation function controls how we'll pass that information on to the next layer. An ideal

Feature selection can be an important part of the machine learning process as it has the ability to greatly improve the performance of our models. While it might seem intuitive to provide your model with all of the information you have with the thinking

Sometimes when we're performing clustering on a dataset, there exist points which don't belong strongly to any given cluster. If we were to use something like k-means clustering [https://www.jeremyjordan.me/grouping-data-points-with-k-means-clustering/], we're forced to make a

This post aims to discuss what a neural network is and how we represent it in a machine learning model. Subsequent posts will cover more advanced topics such as training and optimizing a model, but I've found it's helpful to

Today we'll be talking about support vector machines (SVM); this classifier works well in complicated feature domains, albeit requiring clear separation between classes. SVMs don't work well with noisy data, and the algorithm scales roughly cubic $O(n^3)$ to

Boosting is an iterative ensembling process where models are trained in a sequential order. These models are known as "weak learners" as they are simple prediction rules which only perform slightly better than a random guess (ie. slightly better than 50% accuracy)

Decision trees [https://www.jeremyjordan.me/decision-trees/] are desirable in that they scale well to larger datasets, they are robust against irrelevant features, and it is very easy to visualize the rationalization between a decision tree's predictions. Further, decision trees have low

Many machine learning techniques involve building a model that is capable of representing the data and then finding the optimal parameters for the model to minimize error. K-nearest neighbors, however, is an example of instance-based learning where we instead simply store the training data

Naive Bayes classification methods are quite simple (in terms of model complexity) and commonly used for tasks such as document classification and spam filtering. This algorithm works well for datasets with a large amount of features (ex. a body of text where every word

The goal of logistic regression, as with any classifier, is to figure out some way to split the data to allow for an accurate prediction of a given observation's class using the information present in the features. (For instance, if we were

Decision trees are one of the oldest and most widely-used machine learning models, due to the fact that they work well with noisy or missing data, can easily be ensembled to form more robust predictors, and are incredibly fast at runtime. Moreover, you can

Before you read this post, go ahead and check out my post on decision trees for classification [https://www.jeremyjordan.me/decision-trees-for-classification/]. This post will be building on top of that, as you'll see that decision tree regressors are very similar to

Polynomial regression is very similar to linear regression [https://www.jeremyjordan.me/linear-regression/], with a slight deviation in how we treat our feature-space. Confused? It'll make more sense in a minute, just bear with me. As a reminder, linear regression models are

Before you're ready to feed a dataset into your machine learning model of choice, it's important to do some preprocessing so the data behaves nicely for our model. In an ideal world, you'll have a perfectly clean dataset

Preface: this post is for the complete novice in the field of machine learning in an attempt to orient readers for my other posts on machine learning. This is a high-level view of machine learning. What the heck is machine learning? The whole concept

Linear regression is used to predict an outcome given some input value(s). While machine learning classifiers use features to predict a discrete label for a given instance or example, machine learning regressors have the ability use features to predict a continuous outcome for

In this third post on supervised machine learning classifiers, I'll be talking about one of the oldest and most widely used techniques - decision trees. Decision trees work well with noisy or missing data and are incredibly fast at runtime. They'