-
Notifications
You must be signed in to change notification settings - Fork 30.8k
Make StaticCache configurable at model construct time #32830
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Merged
gante
merged 3 commits into
huggingface:main
from
guangy10:config_model_with_static_cache
Sep 10, 2024
Merged
Changes from all commits
Commits
Show all changes
3 commits
Select commit
Hold shift + click to select a range
File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
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
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
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,33 @@ | ||
<!--Copyright (c) Meta Platforms, Inc. and affiliates. | ||
All rights reserved. | ||
|
||
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with | ||
the License. You may obtain a copy of the License at | ||
|
||
http://www.apache.org/licenses/LICENSE-2.0 | ||
|
||
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on | ||
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the | ||
specific language governing permissions and limitations under the License. | ||
|
||
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be | ||
rendered properly in your Markdown viewer. | ||
|
||
--> | ||
|
||
|
||
# ExecuTorch | ||
|
||
[`ExecuTorch`](https://github.com/pytorch/executorch) is an end-to-end solution for enabling on-device inference capabilities across mobile and edge devices including wearables, embedded devices and microcontrollers. It is part of the PyTorch ecosystem and supports the deployment of PyTorch models with a focus on portability, productivity, and performance. | ||
|
||
ExecuTorch introduces well defined entry points to perform model, device, and/or use-case specific optimizations such as backend delegation, user-defined compiler transformations, memory planning, and more. The first step in preparing a PyTorch model for execution on an edge device using ExecuTorch is to export the model. This is achieved through the use of a PyTorch API called [`torch.export`](https://pytorch.org/docs/stable/export.html). | ||
|
||
|
||
## ExecuTorch Integration | ||
|
||
An integration point is being developed to ensure that 🤗 Transformers can be exported using `torch.export`. The goal of this integration is not only to enable export but also to ensure that the exported artifact can be further lowered and optimized to run efficiently in `ExecuTorch`, particularly for mobile and edge use cases. | ||
|
||
[[autodoc]] integrations.executorch.TorchExportableModuleWithStaticCache | ||
- forward | ||
|
||
[[autodoc]] integrations.executorch.convert_and_export_with_cache |
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
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
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
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
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
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,159 @@ | ||
# Copyright (c) Meta Platforms, Inc. and affiliates. | ||
# All rights reserved. | ||
# | ||
# Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with | ||
# the License. You may obtain a copy of the License at | ||
# | ||
# http://www.apache.org/licenses/LICENSE-2.0 | ||
# | ||
# Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on | ||
# an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the | ||
# specific language governing permissions and limitations under the License. | ||
|
||
import torch | ||
guangy10 marked this conversation as resolved.
Outdated
Show resolved
Hide resolved
|
||
|
||
from transformers import ( | ||
PreTrainedModel, | ||
StaticCache, | ||
) | ||
from transformers.pytorch_utils import is_torch_greater_or_equal_than_2_3 | ||
|
||
|
||
class TorchExportableModuleWithStaticCache(torch.nn.Module): | ||
""" | ||
A wrapper module designed to make a `PreTrainedModel` exportable with `torch.export`, | ||
specifically for use with static caching. This module ensures that the exported model | ||
is compatible with further lowering and execution in `ExecuTorch`. | ||
|
||
Note: | ||
This class is specifically designed to support export process using `torch.export` | ||
in a way that ensures the model can be further lowered and run efficiently in `ExecuTorch`. | ||
""" | ||
|
||
def __init__(self, model: PreTrainedModel): | ||
""" | ||
Initializes the wrapper module with the pretrained model. | ||
|
||
Args: | ||
model (`PreTrainedModel`): The pretrained model to wrap. The model must have caching | ||
enabled and use a 'static' caching implementation. | ||
|
||
Raises: | ||
AssertionError: If the pretrained model does not have caching enabled or if it does | ||
not use a 'static' caching implementation in `model.generation_config`. | ||
""" | ||
super().__init__() | ||
|
||
# Sanity checks | ||
if model.generation_config is None: | ||
raise AssertionError( | ||
"The model must have a generation config to be exported with static caching. " | ||
"Please set `generation_config`." | ||
) | ||
|
||
if not model.generation_config.use_cache: | ||
raise AssertionError( | ||
"The model must have caching enabled to be exported with static caching. " | ||
"Please set `generation_config.use_cache=True`." | ||
) | ||
|
||
if model.generation_config.cache_implementation != "static": | ||
raise AssertionError( | ||
"The model must use a 'static' caching implementation to be exported with static caching. " | ||
"Please set `generation_config.cache_implementation='static'`." | ||
) | ||
|
||
self.model = model | ||
self.static_cache = StaticCache( | ||
config=self.model.config, | ||
batch_size=self.model.generation_config.cache_config.batch_size, | ||
max_cache_len=self.model.generation_config.cache_config.max_cache_len, | ||
dtype=self.model.config.torch_dtype, | ||
) | ||
self.is_causal = any("CausalLM" in arch for arch in self.model.config.architectures) | ||
guangy10 marked this conversation as resolved.
Outdated
Show resolved
Hide resolved
|
||
if self.is_causal: | ||
causal_mask = torch.tril( | ||
torch.ones( | ||
self.static_cache.max_cache_len, | ||
self.static_cache.max_cache_len, | ||
dtype=torch.bool, | ||
) | ||
) | ||
self.register_buffer("mask", causal_mask, persistent=False) | ||
|
||
def forward(self, input_ids: torch.Tensor, cache_position: torch.Tensor): | ||
""" | ||
Forward pass of the module, which is compatible with the ExecuTorch runtime. | ||
|
||
Args: | ||
input_ids (`torch.Tensor`): Tensor representing current input token id to the module. | ||
cache_position (`torch.Tensor`): Tensor representing current input position in the cache. | ||
|
||
Returns: | ||
torch.Tensor: Logits output from the model. | ||
|
||
This forward adapter serves two primary purposes: | ||
|
||
1. **Making the Model `torch.export`-Compatible**: | ||
The adapter hides unsupported objects, such as the `Cache`, from the graph inputs and outputs, | ||
enabling the model to be exportable using `torch.export` without encountering issues. | ||
|
||
2. **Ensuring Compatibility with `ExecuTorch` runtime**: | ||
The adapter matches the model's forward signature with that in `executorch/extension/llm/runner`, | ||
ensuring that the exported model can be executed in `ExecuTorch` out-of-the-box. | ||
""" | ||
_, seqlen = input_ids.shape | ||
attn_mask = self.mask[cache_position, :seqlen] if self.is_causal else None | ||
guangy10 marked this conversation as resolved.
Outdated
Show resolved
Hide resolved
guangy10 marked this conversation as resolved.
Outdated
Show resolved
Hide resolved
|
||
outs = self.model( | ||
input_ids=input_ids, | ||
attention_mask=attn_mask, | ||
position_ids=cache_position.unsqueeze(0), | ||
cache_position=cache_position, | ||
past_key_values=self.static_cache, | ||
use_cache=True, | ||
) | ||
return outs.logits | ||
|
||
|
||
def convert_and_export_with_cache( | ||
model: PreTrainedModel, | ||
example_input_ids: torch.Tensor = None, | ||
example_cache_position: torch.Tensor = None, | ||
): | ||
""" | ||
Convert a `PreTrainedModel` into an exportable module and export it using `torch.export`, | ||
ensuring the exported model is compatible with `ExecuTorch`. | ||
|
||
Args: | ||
model (`PreTrainedModel`): The pretrained model to be exported. | ||
example_input_ids (`torch.Tensor`): Example input token id used by `torch.export`. | ||
example_cache_position (`torch.Tensor`): Example current cache position used by `torch.export`. | ||
|
||
Returns: | ||
Exported program (`torch.export.ExportedProgram`): The exported program generated via `torch.export`. | ||
""" | ||
|
||
if not is_torch_greater_or_equal_than_2_3: | ||
raise ImportError("torch >= 2.3 is required.") | ||
|
||
import torch.export._trace | ||
|
||
with torch.no_grad(): | ||
# TODO: The default inputs only work for text models. We need to add support for vision/audio models. | ||
example_input_ids = ( | ||
example_input_ids if example_input_ids is not None else torch.tensor([[1]], dtype=torch.long) | ||
) | ||
example_cache_position = ( | ||
example_cache_position if example_cache_position is not None else torch.tensor([0], dtype=torch.long) | ||
) | ||
guangy10 marked this conversation as resolved.
Outdated
Show resolved
Hide resolved
|
||
|
||
# Due to issue https://github.com/pytorch/pytorch/issues/128394, we need to switch to use an internal | ||
# export API and pre_dispatch=False. Switch to use the public API once the issue is included in 2.5 release. | ||
exported_program = torch.export._trace._export( | ||
TorchExportableModuleWithStaticCache(model), | ||
args=(example_input_ids,), | ||
kwargs={"cache_position": example_cache_position}, | ||
pre_dispatch=False, | ||
strict=True, | ||
) | ||
return exported_program |
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
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
Oops, something went wrong.
Oops, something went wrong.
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.
Uh oh!
There was an error while loading. Please reload this page.