Skip to content

Features/add stream support#58

Open
Mistobaan wants to merge 2 commits into
anhaidgroup:masterfrom
Unifyd-Insights:features/add-stream-support
Open

Features/add stream support#58
Mistobaan wants to merge 2 commits into
anhaidgroup:masterfrom
Unifyd-Insights:features/add-stream-support

Conversation

@Mistobaan
Copy link
Copy Markdown

planning on contributing few more patches. It would be useful to have a standard format.
right now I used the pep8 with yapf and pylintrc (the standard tensorflow rc).

other patches contain fixes for lazy loading (ie. no fixed .format) and for loading the datasets from a stream (used in online batch setting).

Let me know if these are welcome and I will try to propose them as I work on my own repository.

@sidharthms
Copy link
Copy Markdown
Collaborator

sidharthms commented May 22, 2020 via email

@Mistobaan Mistobaan force-pushed the features/add-stream-support branch from 0aac300 to 50192cd Compare May 27, 2020 08:19
@sidharthms
Copy link
Copy Markdown
Collaborator

It seems using the TF formatting config is causing a lot of changes - probably because of the 80 character limit that Google uses. It's kinda hard to isolate the actual changes in this PR. It would be awesome if you could just have the main changes and keep the formatting as is for the unchanged parts - I believe the default yapf formatting (probably with a line limit of 120) should result in around the same formatting as the code currently has.

class CacheStaleException(Exception):
r"""Raised when the dataset cache is stale and no fallback behavior is specified.
"""
pass
Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why is this line deleted?

begin = timer()
dataset_args = {'fields': fields, 'column_naming': column_naming}
dataset = MatchingDataset(path=path, **dataset_args)
dataset = MatchingDataset(stream=stream, **dataset_args)
Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Maybe I'm missing something - MatchingDataset does not take "stream" as argument right?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants