Overview
Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Learn how to approach the Bengali.AI Handwritten Grapheme Classification Kaggle competition using PyTorch in this comprehensive tutorial video. Dive into a multi-label classification challenge involving over 200,000 images as you build a framework, create a dataset class, and explore image data. Master techniques for training pre-trained models on this specific dataset, covering topics such as workspace setup, data downloading, multilabel classification, code execution, data importing, image pickles, dataset class creation, augmentation application, and dataset verification. Gain valuable insights into tackling complex machine learning competitions and enhancing your PyTorch skills.
Syllabus
Introduction
Setup Workspace
Download Data
Multilabel Classification
Running the Code
Importing Data
Image Pickles
Dataset Class
Apply Augmentation
Verify Datasets
Taught by
Abhishek Thakur