saymrwulf/onnxruntime
1
0
Fork 0
mirror of https://github.com/saymrwulf/onnxruntime.git synced 2026-07-04 04:07:22 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Fix argparser in matmul_bnb4_quantizer (#19812)

Browse source 
### Description
<!-- Describe your changes. -->
The argparser had incorrectly used `description` and `options` instead
of `help` and `choices`.


### Motivation and Context
<!-- - Why is this change required? What problem does it solve?
- If it fixes an open issue, please link to the issue here. -->
Fixes:
#19751
This commit is contained in:
Jambay Kinley 2024-03-07 11:31:34 -08:00 committed by GitHub
parent 72ce4de07d
commit 3dfce2f1cd
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 2 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
onnxruntime/python/tools/quantization/matmul_bnb4_quantizer.py
View file

@ -199,14 +199,14 @@ into a set of 4b integers with an absolute value scaling factor.
"--quant_type",
required=False,
default=1,
options=[MatMulBnb4Quantizer.FP4, MatMulBnb4Quantizer.NF4],
choices=[MatMulBnb4Quantizer.FP4, MatMulBnb4Quantizer.NF4],
help="Quantization data type. 0: FP4, 1: NF4",
)
parser.add_argument(
"--block_size",
required=False,
default=64,
description="Block size for blockwise quantization. Note: bnb.nn.Linear4bit only uses block_size=64",
help="Block size for blockwise quantization. Note: bnb.nn.Linear4bit only uses block_size=64",
)
parser.add_argument("-v", "--verbose", required=False, action="store_true")
parser.set_defaults(verbose=False)