Skip to content
← All tracks

Batching and the Data Path

PyTorch

Stack versus cat, padding and masks, collate functions, and shuffling you can reproduce. The layer where shape bugs are cheapest to prevent and most expensive to find.

0 / 5 solved
Continue →
  1. 1. Not solved yet. Adding an axis, or extending one
  2. 2. Not solved yet. Pad, and then say where you padded
  3. 3. Not solved yet. The mean of the means
  4. 4. Not solved yet. Splitting is free
  5. 5. Not solved yet. Shuffle the same way twice

Check yourself

4 questions · one attempt each

These do not count toward finishing the track. They are here to catch the things that are easy to read past.

0 / 4

GPU utilisation is low and the training loop is waiting. What is num_workers for? 

DataLoader(ds, num_workers=4, pin_memory=True)
Question 1 of 4