Skip to content

[WIP] Reference representation of dqlinear int4 for xnnpack #2507

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.

Sign up for GitHub

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

Closed
wants to merge 1 commit into from
Closed

[WIP] Reference representation of dqlinear int4 for xnnpack #2507

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
227 changes: 227 additions & 0 deletions torchao/quantization/pt2e/reference_representation_rewrite.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,6 +24,9 @@
remove_tensor_overload_for_qdq_ops,
)

from torchao.quantization.quant_primitives import _DTYPE_TO_QVALUE_BOUNDS, MappingType
from torchao.quantization.utils import _get_per_token_block_size

try:
from torch._export.utils import _disable_aten_to_metadata_assertions
except:
Expand All @@ -32,6 +35,8 @@

__all__ = [
"reference_representation_rewrite",
"_qdq_dynamic_quantized_linear_4bit_groupwise",
"_reference_dynamic_quantized_linear_4bit_groupwise",
]


Expand Down Expand Up @@ -203,6 +208,185 @@ def _reference_dynamic_quantized_linear(
return out_fp32


def _qdq_dynamic_quantized_linear_4bit_groupwise(
x_fp32,
x_quant_min,
x_quant_max,
x_eps,
x_scales_type,
x_zero_points_type,
weight_i4,
weight_scale,
weight_zero_point,
weight_quant_min,
weight_quant_max,
bias_fp32,
group_size,
):
# Dynamic quantization of activation
x_mapping_type = MappingType.ASYMMETRIC
per_token_block_size = _get_per_token_block_size(x_fp32)
x_scale, x_zero_point = torch.ops.torchao.choose_qparams_affine(
x_fp32,
x_mapping_type.name,
per_token_block_size,
torch.int8,
x_quant_min,
x_quant_max,
x_eps,
x_scales_type,
x_zero_points_type,
)
x_i8 = torch.ops.torchao.quantize_affine(
x_fp32,
per_token_block_size,
x_scale,
x_zero_point,
torch.int8,
x_quant_min,
x_quant_max,
)
x_fp32 = torch.ops.torchao.dequantize_affine(
x_i8,
per_token_block_size,
x_scale,
x_zero_point,
torch.int8,
x_quant_min,
x_quant_max,
torch.float32,
)

assert group_size > 0, "Group size must be positive"
assert (
weight_i4.shape[1] % group_size == 0
), "Weight must be divisible by group_size"
assert weight_i4.dim() == 2, "Weight must be 2D tensor"
block_size = (1, group_size)
weight_fp32 = torch.ops.torchao.dequantize_affine(
weight_i4,
block_size,
weight_scale,
weight_zero_point,
torch.int8,
weight_quant_min,
weight_quant_max,
torch.float32,
)

out_fp32 = torch.ops.aten.linear.default(x_fp32, weight_fp32, bias_fp32)
return out_fp32


def _reference_dynamic_quantized_linear_4bit_groupwise(
x_fp32,
x_quant_min,
x_quant_max,
x_eps,
x_scales_type,
x_zero_points_type,
weight_i4,
weight_scale,
weight_zero_point, # Not used because assuming weight is symmetric
weight_quant_min,
weight_quant_max,
bias_fp32,
group_size,
):
# Dynamic quantization of activation
x_mapping_type = MappingType.ASYMMETRIC
per_token_block_size = _get_per_token_block_size(x_fp32)
x_scale, x_zero_point = torch.ops.torchao.choose_qparams_affine(
x_fp32,
x_mapping_type.name,
per_token_block_size,
torch.int8,
x_quant_min,
x_quant_max,
x_eps,
x_scales_type,
x_zero_points_type,
)
x_i8 = torch.ops.torchao.quantize_affine(
x_fp32,
per_token_block_size,
x_scale,
x_zero_point,
torch.int8,
x_quant_min,
x_quant_max,
)

# For groupwise quantization, we need to handle the computation differently
# weight_i4 shape: [out_features, in_features]
# weight_scale shape: [out_features, in_features // group_size]
# weight_zero_point shape: [out_features, in_features // group_size]
out_features, in_features = weight_i4.shape
num_groups = in_features // group_size

# scales in xnnpack are stored as bf16 and converted to fp32 for computation
weight_scale = weight_scale.to(torch.bfloat16).to(torch.float32)

assert x_i8.dim() == 2, "x_i8 must be 2D tensor"
# Reshape for group-wise processing
# x: [batch_size, in_features] -> [batch_size, num_groups, group_size]
batch_size = x_i8.shape[0]
x_i8_grouped = x_i8.view(batch_size, num_groups, group_size)

# weight: [out_features, in_features] -> [out_features, num_groups, group_size]
weight_i4_grouped = weight_i4.view(out_features, num_groups, group_size)

# Convert to int16 for computation
x_i32_grouped = x_i8_grouped.to(torch.int32)
weight_i32_grouped = weight_i4_grouped.to(torch.int32)

# Perform groupwise integer linear operation
acc_fp32 = torch.zeros(
batch_size, out_features, dtype=torch.float32, device=x_fp32.device
)

for group_idx in range(num_groups):
# Extract current group
x_group = x_i32_grouped[:, group_idx, :] # [batch_size, group_size]
weight_group = weight_i32_grouped[:, group_idx, :] # [out_features, group_size]
weight_group_col_sum = weight_group.sum(dim=-1) # [out_features]

# Get scale for this group
weight_scale_group = weight_scale[:, group_idx] # [out_features]

# Integer matmul: [batch_size, group_size] @ [group_size, out_features] -> [batch_size, out_features]
group_acc = out_dtype(
torch.ops.aten.linear.default,
torch.int32,
x_group,
weight_group,
None,
)

# Output has to be scaled by x_scale * weight_scale_group
# However we will first scale by weight_scale_group, that is accounting
# only for scale of weight, and then scale by x_scale at the end because
# x_scale applies to all groups
acc_fp32 = acc_fp32 + group_acc.to(torch.float32) * weight_scale_group.view(
1, -1
)

# we must also subtract x_zero_point * weight_group_sum
# since (X - x_zero_point) * W = X * W - x_zero_point * W
weights_col_sum_adjusted = (
weight_group_col_sum.to(torch.float32).view(1, -1)
* x_zero_point.view(-1, 1)
* weight_scale_group.view(1, -1)
)
acc_fp32 = acc_fp32 - weights_col_sum_adjusted
x_scale_multiplier = x_scale.view(-1, 1)
out_fp32 = acc_fp32 * x_scale_multiplier
if bias_fp32 is not None:
out_fp32 = out_fp32 + bias_fp32

return out_fp32


def _qdq_quantized_conv2d(
x_i8,
x_scale,
Expand Down Expand Up @@ -738,6 +922,22 @@ def reference_representation_rewrite(model: GraphModule) -> GraphModule:
127,
)

_DYNAMIC_QUANTIZED_LINEAR_4BIT_GROUPWISE_EXAMPLE_INPUTS = (
torch.randn((2, 32), dtype=torch.float), # x_fp32
-128, # x_quant_min
127, # x_quant_max
torch.finfo(torch.float32).eps, # x_eps
torch.randint(-8, 7, (8, 32), dtype=torch.int8), # weight_i4 (stored as int8)
torch.randn(8, 4, dtype=torch.float), # weight_scale [out_features, num_groups]
torch.zeros(
8, 4, dtype=torch.int
), # weight_zero_point [out_features, num_groups]
torch.tensor([-8], dtype=torch.int), # weight_quant_min (4-bit range)
torch.tensor([7], dtype=torch.int), # weight_quant_max (4-bit range)
torch.randn(8, dtype=torch.float), # bias_fp32
8, # group_size
)

_REWRITE_INFO_LIST = [
_RewriteInfo(
_DYNAMIC_QUANTIZED_LINEAR_EXAMPLE_INPUTS,
Expand All @@ -752,6 +952,33 @@ def reference_representation_rewrite(model: GraphModule) -> GraphModule:
literal_to_ph_idx={-128: 1, 127: 2, torch.finfo(torch.float32).eps: 3},
),
),
_RewriteInfo(
_DYNAMIC_QUANTIZED_LINEAR_4BIT_GROUPWISE_EXAMPLE_INPUTS,
WrapperModule(_qdq_dynamic_quantized_linear_4bit_groupwise),
WrapperModule(_reference_dynamic_quantized_linear_4bit_groupwise),
partial(
_replace_literals_with_existing_placeholders,
literal_to_ph_idx={
-128: 1,
127: 2,
torch.finfo(torch.float32).eps: 3,
-8: 7,
7: 8,
8: 10,
},
),
partial(
_replace_literals_with_existing_placeholders,
literal_to_ph_idx={
-128: 1,
127: 2,
torch.finfo(torch.float32).eps: 3,
-8: 7,
7: 8,
8: 10,
},
),
),
_RewriteInfo(
_QUANTIZED_LINEAR_EXAMPLE_INPUTS,
WrapperModule(_qdq_quantized_linear),
Expand Down
Loading
Loading