Our goals for the homework assignment this week are:
- Practice working with language modeling logits
- Fill in gaps in your PyTorch knowledge
Tutorials
By now we’ve worked with PyTorch a lot, but there are probably some gaps in your knowledge. PyTorch has extensive tutorials on various topics, at various different levels, so everyone should be able to find something to learn there. So: pick two resources from the tutorials, work through them on your own, and write a paragraph about your experience for each. Try to make this something you learn from, not just something you “get through”.
Notice that the left menu on the PyTorch tutorials list has many categories. Pick something from two different categories. I recommend considering:
- PyTorch Recipes: there’s lots of good options here
- Introduction to PyTorch: if you choose this category, consider the entire category to be one of your two resources to work through, though you will probably be able to skim.
- Introduction on YouTube: if you choose this category, type out the code yourself.
- Image and Video: these all look good
- Text: these actually aren’t as good because they use different abstractions than the Hugging Face library.
- RL, Deploying, Parallel, Mobile, and Recommendation are more advanced options that may appeal to some students. (We’ll do a bit of RL later.)
Alternative: If you prefer, you can instead do the classifier-from-scratch mini-project that we made optional back in Unit 6.
This Week’s Fundamentals
There’s only one notebook this week, designed to help you practice with language modeling. As usual, do your work in the corresponding notebook.
Submitting
I recommend working on the other answers outside of Moodle (e.g., in Word or Google Docs), then copy-pasting in.
In the Moodle assignment for this Homework:
- Attach the
ipynbfile. Make sure you’ve done a “Restart and Run All” before submitting. - Copy and paste your responses to the Analysis questions into the text box.