C.W.K.
Stream
Quiz · 4 questions

TPUs & Distributed Training

From 1 GPU to a TPU pod — distribution strategies, mixed precision, scaling recipes

Level 0Level 0
0 XP0/78 lessons0/17 achievements
0/100 XP to next level100 XP to go0% complete

Quiz

01MirroredStrategy with 4 GPUs and per-replica batch 64. What's the global batch size from the optimizer's perspective?
02Why drop_remainder=True for TPU training?
03Correct mixed precision policy for TPU, and why?
04Role of TF_CONFIG in MultiWorkerMirroredStrategy?
Spotted a bug or have feedback on this page?Report an Issue

Comments 0

🔔 Reply notifications (sign in)
Sign inPlease sign in to comment.

No comments yet — be the first.