Add DecomposeGruPass for ARM backend (#17137)#19463
Open
apullin wants to merge 2 commits into
Open
Conversation
added 2 commits
May 11, 2026 11:49
Summary:
Adds quantizable versions of GRU and RNN modules that can be used with
PyTorch quantization-aware training (QAT) for the ARM backend.
The standard nn.GRU and nn.RNN are opaque composite ops that the quantizer
cannot annotate. These modules decompose the RNN operations into
nn.Linear + FloatFunctional so that QAT observers can be inserted at
each arithmetic boundary.
## New modules:
- `GRUCell`, `_GRUSingleLayer`, `_GRULayer`, `GRU`
- `RNNCell`, `_RNNSingleLayer`, `_RNNLayer`, `RNN`
## Features:
- `from_float()` class method to convert from nn.GRU/nn.RNN
- Multi-layer support
- Bidirectional support
- Both tanh and relu nonlinearities (for RNN)
## Usage:
```python
from executorch.backends.arm.quantizable import GRU, RNN
# Create quantizable GRU
model = GRU(input_size=10, hidden_size=20, num_layers=2)
# Or convert from existing nn.GRU
eager_model = torch.nn.GRU(10, 20, 2)
eager_model.qconfig = torch.ao.quantization.get_default_qat_qconfig("fbgemm")
quantizable_model = GRU.from_float(eager_model)
```
Differential Revision: D92059608
Summary:
Adds a decomposition pass that transforms aten.gru.input into elementary
ops supported by TOSA (matmul, sigmoid, tanh, mul, add, slice, cat).
GRU cell equations per timestep:
r_t = sigmoid(x_t @ W_ir.T + b_ir + h_{t-1} @ W_hr.T + b_hr)
z_t = sigmoid(x_t @ W_iz.T + b_iz + h_{t-1} @ W_hz.T + b_hz)
n_t = tanh(x_t @ W_in.T + b_in + r_t * (h_{t-1} @ W_hn.T + b_hn))
h_t = n_t + z_t * (h_{t-1} - n_t)
Features:
- Multi-layer GRU support
- Bidirectional GRU support
- With/without bias
- batch_first support
- Batched gate computation (2 mm ops per timestep instead of 6)
Differential Revision: D92058313
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/19463
Note: Links to docs will display an error until the docs builds have been completed. ❌ 6 New Failures, 1 Unrelated FailureAs of commit 60d56eb with merge base 126507c ( NEW FAILURES - The following jobs have failed:
FLAKY - The following job failed but was likely due to flakiness present on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
meta-cla
Bot
added
the
CLA Signed
Contributor
|
|
This PR needs a
|
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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.
Summary:
Adds a decomposition pass that transforms aten.gru.input into elementary
ops supported by TOSA (matmul, sigmoid, tanh, mul, add, slice, cat).
GRU cell equations per timestep:
r_t = sigmoid(x_t @ W_ir.T + b_ir + h_{t-1} @ W_hr.T + b_hr)
z_t = sigmoid(x_t @ W_iz.T + b_iz + h_{t-1} @ W_hz.T + b_hz)
n_t = tanh(x_t @ W_in.T + b_in + r_t * (h_{t-1} @ W_hn.T + b_hn))
h_t = n_t + z_t * (h_{t-1} - n_t)
Features:
Differential Revision: D92058313
cc @digantdesai @freddan80 @per @zingo @oscarandersson8218 @mansnils @Sebastian-Larsson @robell @rascani