Talks & Workshops

Major league hacking show & Tell

An introduction to PyTorch

15 April, 2021  Workshop resources

About the workshop:
The Show & Tell will give a brief introduction to pytorch. Topics discussed are tensors, computational graphs, autograd. Finally, building a linear model and simple neural network to get an understanding of the workflow in PyTorch.


NTU Open Source Society TGIFHacks

Introduction to Natural Language Processing

29 Jan, 2021  Workshop resources

About the workshop:
Natural language processing is a rich domain in the field of deep learning concerning the interaction between computers and human language. It is one of the most broadly applied areas of deep learning with applications such as question-answering, sentiment analysis, machine translation, and chatbots.
In this workshop, we cover fundamentals of NLP such as word embeddings, probabilistic and neural language models. We’ll cover all the theory behind these topics and finally build a simple language model to get some hands-on experience. Understanding these foundational concepts will help you to learn about more advanced topics such as sequence and attention models.
In this workshop, I covered the fundamentals of NLP such as word embeddings, probablistic and neural language models.


NTU Open Source Society TGIFHacks

Data scraping and Data cleaning

25 September, 2020  Workshop resources

About the workshop:
In this age of data-driven world, scraping and crawling content from the web to create datasets is a crucial skill to have in your portfolio. This workshop aims to give attendees a brief introduction into data scraping from web pages to cleaning the scraped data. By the end of the workshop, attendees should have acquired some hands-on experience with the topic by creating their very own datasets.
In this workshop, attendees will learn about the entire data preparation and collection stage in a machine learning pipeline. We’ll be using scrapy (a python web-crawling framework) to scrape content from webpages and use various python libraries to preprocess our data. By the end of the workshop, the attendees will create a news headlines dataset for sentiment analysis task and Paintings Vs Photographs dataset for binary image classification dataset.