The field of machine learning is becoming easier and easier to enter thanks to readily available tools, a wide range of open source datasets, and a community open to sharing ideas and giving advice. Almost everything you need to get started is online; it's just a matter of finding it.
To help entry-level enthusiasts get their head around different ML systems and how to implement them, I've put together some of my favorite machine learning tutorials. All of the following articles provide a brief introduction to the systems being covered, talk you through the cleaning, testing, and implementation process, and also provide links to datasets and Gitub repositories so you can follow the same steps on your own.
1. Transformers in NLP: Creating a Translator Model from Scratch
This detailed guide explores transformer architecture by creating a translator that takes an English sentence and translates it to German. It covers data preprocessing, model training, and wraps things up by looking at the results and what could be done to improve the system. This machine learning tutorial is helpful for those who want to better understand transformer architecture on a practical level.
2. How to Build a Movie Recommendation System
At the heart of many platforms is a recommendation system that pushes content to the user by learning from their past activity, preferences, and similarity to other users. This is true of Facebook, Amazon, and Netflix, to name just a few examples. In this machine learning tutorial, you'll learn how to build a recommendation system that utilizes content-based and collaborative filtering systems with an open source dataset of movies and users.
3. Building a Scalable AI Chatbot: From Rules to Deep Learning (MLT)
Put together by the folks at Machine Learning Tokyo (MLT), the above video guide is the introduction to a series about developing, training, and implementing an AI chatbot system. It's a massive six part machine learning tutorial that covers using rules and deep learning, data preprocessing, practical challenges, maintaining conversational context, and more.
4. Create an End to End Object Detection Pipeline Using YOLOV5
This is a handy resource for anyone creating an object detection system for a unique dataset. In this machine learning tutorial, you'll get a chance to learn about creating a dataset, annotating it, creating a project structure, and then training YOLOv5 to detect the images in your dataset. All the annotated data and code is available if you want to follow along before diving into your own custom project.
5. How to Generate Anime Faces Using GANs via Pytorch
This guide gives you a brief introduction to how generative adversarial networks (GANs) are used to generate images, and then guides you through how you can apply them yourself. To do so, the author takes an open source anime face dataset and creates a GAN that generates unique images. The results are a great example of how quickly a basic GANs system can be implemented for image creation.
6. Face_recognition
This guide will help you to understand and build a facial recognition system with Python. The goal of the system is to recognize faces in images, as well as to find facial features such as a person's eyes, nose, and mouth. The guide itself is a starting point, and you'll also find a host of other resources to help you better understand and expand on the capabilities of facial recognition systems.
7. Using Natural Language Processing for Spam Detection
If you're looking for a basic introduction to classification systems, this tutorial will walk you through the creation of a basic binary classification system for spam detection. It will talk you through cleaning and tokenizing your data, selecting a model, and implementing it. You'll also be able to access the code and datasets, and the author is happy to help interested readers through Twitter and LinkedIn.
8. Machine Learning Music Generation through Deep Neural Networks
This tutorial looks at deep learning but from an interesting creative angle: how to build a deep learning model for music creation. To help you through the process, this guide starts with an open-source dataset, then goes through the models used to create the system, and shares a few short examples of AI-created music. I enjoyed this machine learning tutorial because the same techniques can be applied to other musical datasets with some tinkering.
9. How to Automate Surveillance Easily with Deep Learning
The first paragraph of this guide to automating surveillance really sums it up nicely: "This article is a quick tutorial for implementing a surveillance system using object detection based on deep learning. It also compares the performance of different Object Detection models using GPU multiprocessing for inference, on Pedestrian Detection."
10. How to Use Customer Data for Data Predictions
This machine learning tutorial covers a well-known ML use case: data prediction for business. It starts by exploring Starbucks cafe data, then uses the information within the dataset to build a model that predicts whether users who enter a store do so due to having received an offer. It offers a quick introduction to data analysis for a specific task, while also exploring what data is important for these kinds of projects.
---
If you enjoyed this article, follow me on Twitter for similar articles and news. And if you have any favorite machine learning tutorials of your own, get in touch and let me know; I'm excited to read more of what's out there and discover other helpful resources.