[PT2] FBC #3258
Merged
[PT2] FBC #3258
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
github-actions
bot
added
NNCF PT
alexsu52
requested changes
Feb 7, 2025
| weight_port_ids = [3] | ||
| bias_port_id = 4 | ||
|
|
||
| if is_experimental_torch_tracing_enabled(): |
There was a problem hiding this comment.
I'm confuse that the same function from the PyTorch has different signature depends on the trace method. Maybe this is different function and you should reuse the approach from #3237.
There was a problem hiding this comment.
Patched function traced only torch.nn.functional.batch_norm (that inside call torch.batch_norm, that have differ signature), that works only because in models most other used batchnorm as module nn.BatchNorm2d, that call used torch.nn.functional.batch_norm.
In experimental traced only torch.batch_norm.
There was a problem hiding this comment.
@anzr299, could handle this case in your PR?
There was a problem hiding this comment.
Yes, I can handle this case in my PR
nncf/experimental/torch2/function_hook/nncf_graph/nncf_graph_builder.py
Outdated
Show resolved
Hide resolved
tests/torch2/function_hook/quantization/test_fast_bias_correction.py
Outdated
Show resolved
Hide resolved
AlexanderDokuchaev
force-pushed
the
ad/pt2_fbc
branch
from
b83c10b to
6b1e982
Compare
|
daniil-lyakhov
approved these changes
Feb 10, 2025
alexsu52
reviewed
Feb 11, 2025
| weight_port_ids = [3] | ||
| bias_port_id = 4 | ||
|
|
||
| if is_experimental_torch_tracing_enabled(): |
There was a problem hiding this comment.
@anzr299, could handle this case in your PR?
alexsu52
approved these changes
Feb 13, 2025
shumaari
pushed a commit
to shumaari/nncf
that referenced
this pull request
Feb 17, 2025
### Changes - Impanated FBC for experimental tracing - Save graph to GraphModelWrapper, to using graph like for NNCFNetwork - Add ConstantLayerAttributes to constant node - Check is_experimental_torch_tracing_enabled inside patch_torch_operators, to support [optimum-intel ](https://github.com/huggingface/optimum-intel/blob/f601b8b1fda4477ed9f9e4293cf3c9d5cec4ad1b/optimum/intel/openvino/__init__.py#L49) ### Related tickets 152996 --------- Co-authored-by: Alexander Dokuchaev <[email protected]>
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.
Changes
Related tickets
152996