- Tried to learn the fundamentals of language models- BERT, roBERTa, and distilBERT.
- Tried to understand the theory behind BERT architecture.
Tools: Simple Transformers Library, Selenium Webdriver, Numpy ,Pandas ,Matplotlib, Scikit, Wordcloud, NLTK, requests, genism, seaborn
Soft skills: Updated my Trello list
- Caleb shared with me the data he scraped from Drowned with sound (13,000 posts), I’ll be working with this dataset, and probably revisit previous modules to preprocess/explore it.
- I ran into some issues with the Simple Transformers Library and my Python version. I had similar issues with genism earlier. It’s clearly an issue of version control, so I think I have to begin looking into Docker, which has tighter control over versions/libraries.