Multi-Label Classification on Unhealthy Comments - Finetuning RoBERTa with PyTorch - Coding Tutorial
Offered By: rupert ai via YouTube
Course Description
Overview
Learn to fine-tune RoBERTa for multi-label classification of unhealthy comments using PyTorch Lightning in this comprehensive coding tutorial. Explore practical Python implementation techniques to create a language model capable of identifying attributes like sarcasm, hostility, and dismissiveness in online comments. Follow along with step-by-step guidance on setting up Google Colab, importing necessary libraries, inspecting data, creating PyTorch datasets and data modules, building the classifier model, and training and evaluating the results. Access the provided Colab notebook to practice hands-on and gain insights into Hugging Face and Transformer models, focusing on implementation rather than theory. Benefit from additional resources, including research papers on RoBERTa and the Unhealthy Comment Corpus, to deepen your understanding of the underlying concepts.
Syllabus
Intro:
Video / project outline:
Getting Google Colab set up:
Imports:
Inspect data:
Pytorch dataset:
Pytorch lightning data module:
Creating the model / classifier:
Training and evaluating model:
Taught by
rupert ai
Related Courses
Hugging Face Transformers - The Basics - Practical Coding Guides - NLP Models (BERT/RoBERTa)rupert ai via YouTube Programming Language of the Future: AI in Your Native Language
Linux Foundation via YouTube Pre-training and Pre-trained Models in Advanced NLP - Lecture 5
Graham Neubig via YouTube Fine-tuning LLMs Without Maxing Out Your GPU - LoRA for Parameter-Efficient Training
Data Centric via YouTube MLOps: OpenVino Quantized Pipeline for Grammatical Error Correction
The Machine Learning Engineer via YouTube