Status Updates From Building Machine Learning P...
Building Machine Learning Pipelines: Automating Model Life Cycles with TensorFlow by
Status Updates Showing 1-25 of 25
Xianshun Chen
is finished
pretty disappointing, not much value for me personally, most of the features in such a pipeline we already built on our own, we have even more specific features. what's worse, the TFX changes so much since the book's publication, lack of documentation, lack of working code
— Feb 07, 2021 06:49PM
Add a comment
Xianshun Chen
is 30% done
The source code in the book contains lots of errors. also the author's github repo contains only a few samples from the books
— Feb 07, 2021 06:29PM
Add a comment
Xianshun Chen
is 15% done
The source codes in the book contains quite a number of error and cannot be run
— Feb 07, 2021 06:05PM
Add a comment
Xianshun Chen
is 15% done
the source code in the book really needs some revision, example_gen is mentioned without even the import statementm, also bad syntax highlight, cannot tell output from code when given example
— Feb 07, 2021 05:02PM
Add a comment
Xianshun Chen
is 15% done
the source code in the book really needs some revision, example_gen is mentioned without even the import statement
— Feb 07, 2021 03:52PM
Add a comment
Xianshun Chen
is 10% done
not particularly impressed by tfx at this point, the code was broken due to version update, and the api change so significantly, that due to lack of latest documentation on the basic examples, it is literally not runnable, i can either revert to 0.22 (which i don't like to do) or google search for latest example, which difficult to come by considering change in api and debug on my own (yes reverse engineer and guess
— Feb 07, 2021 12:13PM
Add a comment
Vishwanath
is 30% done
text seems reasonably well researched. Github repo accompanying the book is poorly maintained, missing chapters and has notebook names like 'untitled.ipynb'. Few data files and accompanying code seems to be an afterthought. Plan reading another chapter before canning this. Seems like a GCP marketing artifact.
— Aug 11, 2020 03:15PM
Add a comment





