3.x SQ supports calib_func for auto-tune #1812
Closed
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
Signed-off-by: Cheng, Zixuan <[email protected]>
Signed-off-by: Cheng, Zixuan <[email protected]>
Signed-off-by: Cheng, Zixuan <[email protected]>
🌩️ Required checks status: Pending 🟡Groups summary🟡 Code Scan Tests workflow
These checks are required after the changes to 🟡 Model Tests 3x workflow
These checks are required after the changes to 🟡 Unit Tests 3x-PyTorch workflow
These checks are required after the changes to Thank you for your contribution! 💜
|
xin3he
requested changes
May 24, 2024
| @@ -465,6 +473,9 @@ def forward_wrapper(model, input, device=torch.device("cpu")): # pragma: no cov | |||
| output = model(*input) | |||
| except: | |||
| output = model(input) | |||
| elif isinstance(input, zip): | |||
| for args, kwargs in input: | |||
| output = model(*args, **kwargs) | |||
|
|
||
| block_modules = {} | ||
| for key in self.block_names: | ||
| block_modules[key] = get_module(self.model, key) | ||
| self._add_blockwise_observer(block_modules) | ||
|
|
||
| forward_wrapper(self.model, input, self.device) ##disable quant and get fp32 output | ||
| # get input args and kwargs for the first block, then do forward | ||
| total_block_args, total_block_kwargs = get_hidden_states(self.model, calib_sample_num, calib_func) |
There was a problem hiding this comment.
seems like we fused all output into once instead of per input? Will it impact the final result @yintong-lu
|
See new design in #1821 |
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Outdated suggestions cannot be applied.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.
Type of Change
sq supports calib_func for auto-tune, no need for dataloader
Description
Layer-wise & block-wise enable
Add ut check auto-tune
Check llm examples
Expected Behavior & Potential Risk
How has this PR been tested?
Dependency Change?