TPU Speed Data Pipelines Reviews

Mikael T. · Reviewed больше 2 лет ago

Ehsan u. · Reviewed больше 2 лет ago

Nikolas N. · Reviewed больше 2 лет 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 больше 2 лет ago

aishwarya b. · Reviewed больше 2 лет ago

Joost B. · Reviewed больше 2 лет ago

Thorvald Sahl J. · Reviewed больше 2 лет 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 больше 2 лет ago

Aditya S. · Reviewed больше 2 лет ago

Brody I. · Reviewed больше 2 лет 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 больше 2 лет ago

LUCAS S. · Reviewed больше 2 лет ago

Sudharshan N. · Reviewed больше 2 лет ago

Alexey T. · Reviewed больше 2 лет ago

Suchita B. · Reviewed больше 2 лет ago

DANIEL M. · Reviewed больше 2 лет ago

Jayesh Arun B. · Reviewed больше 2 лет ago

ROBERTO O. · Reviewed больше 2 лет ago

Rahul M. · Reviewed больше 2 лет ago

David M. · Reviewed больше 2 лет ago

Harold H. · Reviewed больше 2 лет 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 больше 2 лет ago

Ming H. · Reviewed больше 2 лет ago

LUIS F. · Reviewed больше 2 лет ago

Kidus B. · Reviewed больше 2 лет ago