TPU Speed Data Pipelines Reviews

Nikolas N. · Reviewed neredeyse 2 yıl 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 neredeyse 2 yıl ago

aishwarya b. · Reviewed neredeyse 2 yıl ago

Joost B. · Reviewed neredeyse 2 yıl ago

Thorvald Sahl J. · Reviewed neredeyse 2 yıl 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 neredeyse 2 yıl ago

Aditya S. · Reviewed neredeyse 2 yıl ago

Brody I. · Reviewed neredeyse 2 yıl 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 neredeyse 2 yıl ago

LUCAS S. · Reviewed neredeyse 2 yıl ago

Sudharshan N. · Reviewed neredeyse 2 yıl ago

Alexey T. · Reviewed neredeyse 2 yıl ago

Suchita B. · Reviewed neredeyse 2 yıl ago

DANIEL M. · Reviewed neredeyse 2 yıl ago

Jayesh Arun B. · Reviewed neredeyse 2 yıl ago

ROBERTO O. · Reviewed neredeyse 2 yıl ago

Rahul M. · Reviewed neredeyse 2 yıl ago

David M. · Reviewed neredeyse 2 yıl ago

Harold H. · Reviewed neredeyse 2 yıl 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 neredeyse 2 yıl ago

Ming H. · Reviewed neredeyse 2 yıl ago

LUIS F. · Reviewed neredeyse 2 yıl ago

Kidus B. · Reviewed neredeyse 2 yıl ago

Menegozzo G. · Reviewed neredeyse 2 yıl ago

Marián Ž. · Reviewed neredeyse 2 yıl ago