[Quant][Kernel] Remove GPTQ Group/Dynamic Activation Ordering (#54809)
Signed-off-by: Roderick-Wu <roderickwu2003@gmail.com> Signed-off-by: Roderick Wu <roderickwu2003@gmail.com> Signed-off-by: mgoin <mgoin64@gmail.com> Co-authored-by: mgoin <mgoin64@gmail.com>
This commit is contained in:
@@ -70,7 +70,7 @@ def test_gptq_allspark_gemm_ampere(mnk_factors, group_size, has_zp, dtype):
|
||||
input = rand_data((m, k), dtype=dtype)
|
||||
weight = rand_data((k, n), dtype=dtype)
|
||||
|
||||
# Quantize (and apply act_order if provided)
|
||||
# Quantize weights.
|
||||
w_ref, qw, s, zp = quantize_weights(
|
||||
weight, scalar_types.uint8b128, group_size, has_zp
|
||||
)
|
||||
|
||||
Reference in New Issue
Block a user