TPU Speed Data Pipelines Reviews

Mikael T. · Reviewed about 2 years ago

Ehsan u. · Reviewed about 2 years ago

Nikolas N. · Reviewed about 2 years ago

Instructions again make absolutely no sense - "iterate the files" is not an instruction that is complete is any way. Even the question numbers are wrong just listing question 1 repeatedly.

Oliver H. · Reviewed about 2 years ago

aishwarya b. · Reviewed about 2 years ago

Joost B. · Reviewed about 2 years ago

Thorvald Sahl J. · Reviewed about 2 years ago

It's kind of messy. We're not using accelerators (GPUs) and we're not using parallelism at all. I guess it's nice to learn the data loading.

Loading A. · Reviewed about 2 years ago

Aditya S. · Reviewed about 2 years ago

Brody I. · Reviewed about 2 years ago

Decent but there's a lot missing. TFRecord is introduced and then there is no explanation of tf.io.*, so you'll end up searching through a bunch of docs until maybe you come across parse_single_example or until you look at the solution.

chris w. · Reviewed about 2 years ago

LUCAS S. · Reviewed about 2 years ago

Sudharshan N. · Reviewed about 2 years ago

Alexey T. · Reviewed about 2 years ago

Suchita B. · Reviewed about 2 years ago

DANIEL M. · Reviewed about 2 years ago

Jayesh Arun B. · Reviewed about 2 years ago

ROBERTO O. · Reviewed about 2 years ago

Rahul M. · Reviewed about 2 years ago

David M. · Reviewed about 2 years ago

Harold H. · Reviewed about 2 years ago

2. To use TFRecord format to load training data efficiently from GCS it is more efficient is it because that it doesn't show the pictures, but it shows 000100 010000 01010?

Wanda W. · Reviewed about 2 years ago

Ming H. · Reviewed about 2 years ago

LUIS F. · Reviewed about 2 years ago

Kidus B. · Reviewed about 2 years ago