[camb]add w8a8 support #176
Merged
Conversation
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
jinminxi104
requested changes
Feb 26, 2025
dlinfer/vendor/camb/camb_ops.py
Outdated
| we need to reshape the input tensor to 2D tensor if it is 3D tensor. | ||
| """ | ||
| bsz, seq_len = None, None | ||
| if x.dim() == 3: |
There was a problem hiding this comment.
please merge dim==3 and dim==2. just reshape to (bsz * seq_len, -1) for both dim3 and 2
dlinfer/vendor/camb/camb_ops.py
Outdated
| assert quant_dtype == torch.int8 | ||
| assert quant_granularity == "PER_TOKEN" | ||
| if x.shape[-1] in smooth_dic: | ||
| smooth = smooth_dic[x.shape[-1]] |
There was a problem hiding this comment.
could u explain more about smooth_dic
jinminxi104
requested changes
Feb 26, 2025
dlinfer/vendor/camb/__init__.py
Outdated
| @@ -1,3 +1,6 @@ | |||
| import torch | |||
|
|
|||
| from . import pytorch_patch, camb_ops | |||
|
|
|||
| # TODO. weitao: camb torch-mlu-ops-v1.2.0 per_token_smooth_quantize need smooth_vec | |||
| SMOOTH_VEC = torch.ones(8000, dtype=torch.float32, device="mlu") | |||
There was a problem hiding this comment.
please change to 8*1024
dlinfer/vendor/camb/camb_ops.py
Outdated
| @@ -6,6 +6,8 @@ | |||
| from dlinfer.utils.registry import register_ops | |||
| from dlinfer.utils.type_annotation import Tensor, Optional, Sequence, Tuple | |||
|
|
|||
| from .__init__ import SMOOTH_VEC | |||
There was a problem hiding this comment.
from .__init__ ==> from .
dlinfer/vendor/camb/camb_ops.py
Outdated
| if x.shape[-1] <= SMOOTH_VEC.shape[0]: | ||
| smooth = SMOOTH_VEC[: x.shape[-1]] | ||
| else: | ||
| smooth = torch.ones(x.shape[-1], dtype=torch.float32, device=x.device) |
There was a problem hiding this comment.
Don't create a new one locally, update the global SMOOTH_VEC. and round up to 2^N K.
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.
No description provided.