In this workshop, we will go through the importance of SQL, a programming language that is used to manage relational databases and perform operations on the data.  SQL is a popular language which is used in many relational database management systems. We will explore different SQL techniques, such as querying, filtering, grouping and joining data. Through this workshop, you’ll learn how to obtain mission-critical insights for your analysis on real-world data. 

In this workshop, you will learn how to use the Pandas package on Python, which is a widely used data processing library to manipulate data structures. We will explore different data manipulation techniques, such as how to select and merge data, use apply and statistical functions. You will learn how to manipulate real-world data to obtain new insights and learn what it is like to be a data scientist!

Code 1

Code 2

This workshop provides an Introduction to Machine Learning using Scikit-learn. We dive into the basics and derivation of Machine Learning, data processing as well as how to validate our model. You will learn how to split data into training and testing sets, create and fit a model, followed by improving the model. 

Code

This workshop explores the relationship between statistics and programming. We discuss the theory of key statistics concepts, hypothesis testing models such as T-test and Z-test, and how to use these functions in code. This workshop is a three-part series from data cleaning, to data visualisation using Seaborn and Matplotlib to produce insightful data visualisations, to using statistical tests with Scipy

Code

In this workshop, the NUS SDS Workshops team performs exploratory data analysis and preprocesses image datasets in preparation for an end-to-end CV project. Basic knowledge in OpenCV and various feature engineering techniques was taught to participants.

Code

In this collaboration between NUS SDS and NUS IEEE, students at the workshop are taught on how to fine-tune Convolutional Neural Networks (CNNs) for image classification, other techniques such as modifying network architecture and data augmentation to enhance performance, and various metrics to evaluate performance of the model.

Code

In this Workshop which is the final installment to the "Data Science Fundamentals" series, participants are introduced to the concept of model deployment learn of the different cloud services used to do so, such as the Google Cloud. Participants are taught step by step to configure and deploy a Computer Vision model on the cloud as an exercise for them to apply their newly acquired knowledge.

Code

In this Workshop, Participants of Orbital (CP2106) are introduced to the idea of text generation in NLP using both Naïve Bayesian methods and deep learning. Fundamentals of deep learning, including the concepts of neural networks and gradient descent are also covered, with a live demonstration of the content through Colab notebooks.

Code 1

Code 2

In this workshop, Participants of Orbital (CP2106) are given first-hand experience at performing Web Scraping using popular static and dynamic web scraping softwares, BeautifulSoup and Selenium. The scrape results are then transformed into a dataset used for training a model to be deployed in Flask. Similarly, the content is covered and demonstrated through a Colab code-along session.

Code 1

Code 2

This workshop introduces you to the higher level ideas of Machine Learning, and provides a launching pad for you to explore and learn more about machine learning, and how it plays a role in Software. We will learn about the training loop of Deep Learning occurs using Tensorflow and also on ways to collect data for your usecase in Selenium and BeautifulSoup4. 

Wonder how NLP techniques are applied in order to solve real world problems? Join us as we explore a Kaggle Competition Dataset on Sentiment Analysis and understand how these techniques are applied. We will also cover cutting-edge advancements such as Attention Models like BERT that are being applied in chatbots and other such solutions. 

Heard of Natural Language Processing (NLP) but not sure how to get started? Well, we’ve got you covered! In this workshop, we will go through the basics of NLP using Python. We will use popular text processing libraries such as Spacy to encode and process raw textual data to perform tasks such as sentiment analysis.

The Annual Data Science Competition hosted by NUS Statistics and Data Science Society (NUS SDS) was back again for its 2022 iteration. This year’s competition featured a geospatial dataset graciously sponsored by Grab and participants got to work with real-life transport data across countries in the region.

In this workshop, we will go through 2 popular packages in the Tidyverse library: dplyr and ggplot2. dplyr helps to solve the most common data manipulation challenges while ggplot2 assists in the creation of stunning graphics by mapping variables to aesthetics. The Tidyverse library is an indispensable tool for data analysis and we will be here to guide you through interactive code-alongs in this workshop.

Have you ever wondered what deep learning is, and how it works? In this workshop, we will introduce the components of a neural network, understand how a neural network works, and analyse its output. We will also train a model using TensorFlow, which is a framework created by Google for developing Deep Learning models.

In this workshop, we will go through the higher level principles that go behind Robust System Design for MLOps, to bring models from the Jupyter Notebook to Production. We will see how such principles are applied in addressing business needs in Grab-Singtel Digibank and see a Technical Demonstration of such design principles in action. Hear from a up and coming team, how to bring your Machine Learning Models to life and impact and change lives.

At the 2021 Student Life Fair, hear about the events and activities we organise, and experience a workshop trailer on Exploratory Data Analysis in Python.

We explore the high-level ideas of Exploratory Data Analysis in the Data Science Framework, learn statistical tools used in tabular data, and run through an interactive Hands-on-tutorial in Google Colab.

Ever wondered how statisticians / data scientists analyse data and create visually appealing graphs? This workshop serves as an introduction to the most popular data analysis library - tidyverse in R. Do not worry if you have never used R, we are going to guide you from the very basics of R programming to using dplyr for data wrangling followed by creating meaningful visualisations using ggplot2. Empowered with this knowledge, you shall be able to create your own plots and analyse any dataset that interests you. 

Ever wondered how to convert insights into beautiful visuals and messages that will engage your audience? Join us on 27th Jan to see how we can leverage on Gestalt principles to tell the story of your data more effectively. Come join us to learn more about cognitive load, visual tools and how you can put knowledge such as significance tests to real life application. For those who are interested, we also introduce JASP, a software that could be useful to many of you in the future. 

We’ll start with an introduction to classical computer vision and the OpenCV library, where you will learn about image processing techniques, edge detection, template matching and more, followed by deep learning concepts for object detection. You’ll be exposed to a variety of convolutional neural network-based image models that you can choose to use when building your model for the Data Science Competition. Finally, we will elaborate on how you can leverage Transfer Learning to simplify and accelerate the model development process.

In this age of big data, many algorithms are restricted by large dimensional data, commonly known as the “Curse of Dimensionality”. Dimensionality reduction is a class of methods that can alleviate this problem. In this workshop, we introduce a dimensionality reduction algorithm in-depth, Principal Component Analysis (PCA), covering the statistics and linear algebra prerequisites required, the PCA algorithm and what it computes, and situate the algorithm in a diverse range of applications: Machine Learning, Biology and Finance. The workshop is conducted by the Workshops team’s most diverse people: Ang Yi Zhe (Data Science and Analytics), Ang Ming Liang (Computational Biology and Mathematics) and Zheng Peng (Quantitative Finance). 

Have a real-world problem you want to solve with machine learning (ML), but don’t know where to begin? Fret not - In our project-driven workshop, we will cover everything from picking the right ML tools, sourcing or even creating your own dataset, model training, all the way to deploying it on a web server to be viewed as an interactive, demonstrable product. At the end of the workshop, you can expect to take away your very own Face Mask Detector that you can enhance in any direction you might like. Join us on 17 September to build and deploy your face mask detector👍

Excited to learn about how to work with multiple tables and tap into more powerful functions with SQL? Join us, as we learn to write better SQL functions together in this questions-based workshop. Don’t work for data, make SQL and data work for you instead! 

In this workshop, we employ a questions-based approach to learning SQL for data science. We introduce basic relational database architectures and demonstrate basic queries using SELECT, FROM and WHERE statements to retrieve data that is relevant to your needs. Join us as we learn to have fun writing queries to answer important questions in data science! 

At the Student Life Fair, hear about the events and activities we organise, and experience a workshop trailer on machine learning from scratch!

Introduction to PyTorch, applied to computer vision, natural language processing, reinforcement learning and Fastai.

Introduction to joins, subqueries and advanced queries.

Introduction to SQL, SQL data definition and manipulation and SQL queries (DQL).

Hands-on walkthrough of applying machine learning to your data science projects.

This workshop will introduce you to setting up your model as a serviceable REST API on Flask deployed on Google Cloud Platform. We will also introduce you to the concept and applications of Transfer Learning, using the knowledge we gained from the previous Orbital workshop.