Refactor datatype with simpler design #2647
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
blythed
force-pushed
the
refactor/2634/create-inbuilt-datatype
branch
2 times, most recently
from
91c6329 to
a667d6f
Compare
blythed
changed the title
blythed
force-pushed
the
refactor/2634/create-inbuilt-datatype
branch
21 times, most recently
from
fa4881f to
2f516fe
Compare
blythed
force-pushed
the
refactor/2634/create-inbuilt-datatype
branch
2 times, most recently
from
c955598 to
d8c1197
Compare
jieguangzhou
requested changes
Nov 27, 2024
Comment on lines
42
to
45
| _fields = { | ||
| 'object': dill_serializer, | ||
| 'postprocess': dill_serializer, | ||
| 'preprocess': dill_serializer, | ||
| } |
There was a problem hiding this comment.
It should be t.ClassVar. There are similar cases in other areas as well.
There was a problem hiding this comment.
It is ClassVar by default.
blythed
force-pushed
the
refactor/2634/create-inbuilt-datatype
branch
2 times, most recently
from
98bd8d2 to
0302e5b
Compare
jieguangzhou
approved these changes
Nov 27, 2024
kartik4949
requested changes
Nov 27, 2024
| @@ -161,8 +161,7 @@ def _fit_with_dataloaders( | |||
| self.append_metrics(all_metrics) | |||
| self.log(fold='VALID', iteration=iteration, **all_metrics) | |||
| if self.saving_criterion(): | |||
| db.replace(model, upsert=True) | |||
| self.changed.update({'all_metrics', 'optimizer_state'}) | |||
| ) | ||
| old_uuid = info['uuid'] | ||
| except FileNotFoundError: | ||
| pass |
superduper/components/component.py
Outdated
| @@ -328,6 +366,9 @@ def create_jobs( | |||
|
|
|||
| depends = [d for d in attr.depends if d in triggers] | |||
|
|
|||
| # if attr_name == 'validate_in_db': | |||
| # import pdb; pdb.set_trace() | |||
|
|
||
| # :param shape: The shape of the vector | ||
| # :param identifier: The identifier of the vector | ||
| # """ |
superduper/ext/numpy/encoder.py
Outdated
| # :param shape: The shape of the array. | ||
| # :param bytes_encoding: The bytes encoding to use. | ||
| # :param encodable: The encodable to use. | ||
| # """ |
blythed
force-pushed
the
refactor/2634/create-inbuilt-datatype
branch
from
0302e5b to
534442c
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Main changes in
superduper/component/datatype.py.datatype = Datatype('my-dt', encoder=func_1, decoder=func_2)(now just subclass)Encodablebytesencoding to theSchema, disabling inline encoding without aSchemaDocument.encodeoutput since encodables are no longer needed