[RFC] basic model builder infra #156
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
dan-garvey
force-pushed
the
model_builder
branch
2 times, most recently
from
c506747 to
aefc3c5
Compare
The main motivation here is to be able to programatically create hf models using the bare minimum information. not sure if there are any other use cases besides hf. Also sorely lacking in support for anything that doesn't just work with aot.export, multiple inputs, etc. Ideally this will also be the point at which we do any torch ir caching cloudside.
added some tests. Need to find a way to manage the equivalent of torch-mlir heavydeps
The main motivation here is to be able to programatically create hf models using the bare minimum information. not sure if there are any other use cases besides hf. Also sorely lacking in support for anything that doesn't just work with aot.export, multiple inputs, etc. Ideally this will also be the point at which we do any torch ir caching cloudside.
added some tests. Need to find a way to manage the equivalent of torch-mlir heavydeps
dan-garvey
force-pushed
the
model_builder
branch
7 times, most recently
from
91282a7 to
0a475d5
Compare
dan-garvey
force-pushed
the
model_builder
branch
5 times, most recently
from
9e58c24 to
4d8db4c
Compare
|
The scope of this kind of ballooned, and I left python packaging out of it for now. I didnt get to adding a successful test for the TransformerBuilder, and I didnt end up using it. I can remove it from this pr if preferred and add back later if there is a use for it. |
dan-garvey
force-pushed
the
model_builder
branch
from
4d8db4c to
8ecf4ed
Compare
|
Going to keep turbine models tests split, will add test infra in a separate patch |
IanNod
reviewed
Nov 13, 2023
dan-garvey
force-pushed
the
model_builder
branch
3 times, most recently
from
45685d3 to
da92128
Compare
dan-garvey
force-pushed
the
model_builder
branch
from
da92128 to
7752045
Compare
IanNod
approved these changes
Nov 13, 2023
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.
The main motivation here is to be able to programmatically create hf models (for SHARK) using the bare minimum information. And to have a real api instead of importing from examples. I'm not sure if there are any other use cases besides hf. Also sorely lacking in support for anything that doesn't just work with aot.export, multiple inputs, etc. Ideally this will also be the point at which we do any torch ir caching cloudside.