Skip to content

Commit

Permalink
add TorchQuantLinear
Browse files Browse the repository at this point in the history
  • Loading branch information
@ZX-ModelCloud
ZX-ModelCloud committed Dec 4, 2024
1 parent 0538d79 commit c0f8b11
Show file tree
Hide file tree
Showing 3 changed files with 313 additions and 260 deletions.
325 changes: 66 additions & 259 deletions gptqmodel/nn_modules/qlinear/qlinear_cuda.py
View file
Original file line number Diff line number Diff line change
@@ -1,92 +1,42 @@
import math

import numpy as np
import torch
import torch.nn as nn
import transformers
from gptqmodel.nn_modules.qlinear import BaseQuantLinear
from gptqmodel.utils.logger import setup_logger
from gptqmodel.nn_modules.qlinear.qlinear_torch import TorchQuantLinear

logger = setup_logger()

cuda_import_exception = None
try:
import gptqmodel_cuda_64
import gptqmodel_cuda_256

_gptqmodel_cuda_available = True
except ImportError:
except ImportError as e:
cuda_import_exception = e
logger.warning("CUDA extension not installed.")
gptqmodel_cuda_256 = None
gptqmodel_cuda_64 = None
_gptqmodel_cuda_available = False

class CudaQuantLinear(BaseQuantLinear):

class CudaQuantLinear(TorchQuantLinear):
SUPPORTS_BITS = [2, 3, 4, 8]

def __init__(
self,
bits: int,
group_size: int,
sym: bool,
desc_act: bool,
infeatures: int,
outfeatures: int,
bias: bool,
kernel_switch_threshold=128,
weight_dtype=torch.float16,
**kwargs,
self,
bits: int,
group_size: int,
sym: bool,
desc_act: bool,
infeatures: int,
outfeatures: int,
bias: bool,
kernel_switch_threshold=128,
weight_dtype=torch.float16,
**kwargs,
):
super().__init__(bits=bits, group_size=group_size, sym=sym, desc_act=desc_act, infeatures=infeatures, outfeatures=outfeatures, **kwargs)
global _gptqmodel_cuda_available

self.infeatures = infeatures
self.outfeatures = outfeatures
self.bits = bits
self.group_size = group_size if group_size != -1 else infeatures
self.maxq = 2**self.bits - 1

self.register_buffer(
"qweight",
torch.zeros((infeatures // 32 * self.bits, outfeatures), dtype=torch.int32),
)
self.register_buffer(
"qzeros",
torch.zeros(
(
math.ceil(infeatures / self.group_size),
outfeatures // 32 * self.bits,
),
dtype=torch.int32,
),
)
self.register_buffer(
"scales",
torch.zeros(
(math.ceil(infeatures / self.group_size), outfeatures),
dtype=weight_dtype,
),
)
self.register_buffer(
"g_idx",
torch.tensor([i // self.group_size for i in range(infeatures)], dtype=torch.int32),
)
if bias:
self.register_buffer("bias", torch.zeros((outfeatures), dtype=weight_dtype))
else:
self.bias = None

# is performed by unpacking the weights and using torch.matmul
if self.bits in [2, 4, 8]:
self.wf = torch.tensor(list(range(0, 32, self.bits)), dtype=torch.int32).unsqueeze(0)
elif self.bits == 3:
self.wf = torch.tensor(
[
[0, 3, 6, 9, 12, 15, 18, 21, 24, 27, 30, 0],
[0, 1, 4, 7, 10, 13, 16, 19, 22, 25, 28, 31],
[0, 2, 5, 8, 11, 14, 17, 20, 23, 26, 29, 0],
],
dtype=torch.int32,
).reshape(1, 3, 12)
super().__init__(bits=bits, group_size=group_size, sym=sym, desc_act=desc_act, infeatures=infeatures,
outfeatures=outfeatures, **kwargs)

self.kernel_switch_threshold = kernel_switch_threshold
self.gptqmodel_cuda_available = _gptqmodel_cuda_available
Expand All @@ -97,204 +47,61 @@ def __init__(
if infeatures % 64 != 0 or outfeatures % 64 != 0:
self.gptqmodel_cuda_available = False

def post_init(self):
pass

def pack(self, linear, scales, zeros, g_idx=None):
W = linear.weight.data.clone()
if isinstance(linear, nn.Conv2d):
W = W.flatten(1)
if isinstance(linear, transformers.pytorch_utils.Conv1D):
W = W.t()

self.g_idx = g_idx.clone() if g_idx is not None else self.g_idx

scales = scales.t().contiguous()
zeros = zeros.t().contiguous()
scale_zeros = zeros * scales
self.scales = scales.clone().to(dtype=linear.weight.dtype)
if linear.bias is not None:
self.bias = linear.bias.clone().to(dtype=linear.weight.dtype)

intweight = []
for idx in range(self.infeatures):
intweight.append(
torch.round((W[:, idx] + scale_zeros[self.g_idx[idx]]) / self.scales[self.g_idx[idx]]).to(torch.int)[
:, None
]
)
intweight = torch.cat(intweight, dim=1)
intweight = intweight.t().contiguous()
intweight = intweight.numpy().astype(np.uint32)

i = 0
row = 0
qweight = np.zeros((intweight.shape[0] // 32 * self.bits, intweight.shape[1]), dtype=np.uint32)
while row < qweight.shape[0]:
if self.bits in [2, 4, 8]:
for j in range(i, i + (32 // self.bits)):
qweight[row] |= intweight[j] << (self.bits * (j - i))
i += 32 // self.bits
row += 1
elif self.bits == 3:
for j in range(i, i + 10):
qweight[row] |= intweight[j] << (3 * (j - i))
i += 10
qweight[row] |= intweight[i] << 30
row += 1
qweight[row] |= (intweight[i] >> 2) & 1
i += 1
for j in range(i, i + 10):
qweight[row] |= intweight[j] << (3 * (j - i) + 1)
i += 10
qweight[row] |= intweight[i] << 31
row += 1
qweight[row] |= (intweight[i] >> 1) & 0x3
i += 1
for j in range(i, i + 10):
qweight[row] |= intweight[j] << (3 * (j - i) + 2)
i += 10
row += 1

qweight = qweight.astype(np.int32)
self.qweight = torch.from_numpy(qweight)

zeros = zeros.numpy().astype(np.uint32)
qzeros = np.zeros((zeros.shape[0], zeros.shape[1] // 32 * self.bits), dtype=np.uint32)
i = 0
col = 0
while col < qzeros.shape[1]:
if self.bits in [2, 4, 8]:
for j in range(i, i + (32 // self.bits)):
qzeros[:, col] |= zeros[:, j] << (self.bits * (j - i))
i += 32 // self.bits
col += 1
elif self.bits == 3:
for j in range(i, i + 10):
qzeros[:, col] |= zeros[:, j] << (3 * (j - i))
i += 10
qzeros[:, col] |= zeros[:, i] << 30
col += 1
qzeros[:, col] |= (zeros[:, i] >> 2) & 1
i += 1
for j in range(i, i + 10):
qzeros[:, col] |= zeros[:, j] << (3 * (j - i) + 1)
i += 10
qzeros[:, col] |= zeros[:, i] << 31
col += 1
qzeros[:, col] |= (zeros[:, i] >> 1) & 0x3
i += 1
for j in range(i, i + 10):
qzeros[:, col] |= zeros[:, j] << (3 * (j - i) + 2)
i += 10
col += 1

qzeros = qzeros.astype(np.int32)
self.qzeros = torch.from_numpy(qzeros)

def forward(self, x: torch.Tensor):
out_shape = x.shape[:-1] + (self.outfeatures,)
x = x.reshape(-1, x.shape[-1])
x_dtype = x.dtype
if (
x.device.type == "cuda"
and self.gptqmodel_cuda_available
and (self.kernel_switch_threshold == 0 or x.shape[0] < self.kernel_switch_threshold)
):
out = torch.zeros((x.shape[0], self.outfeatures), device=x.device, dtype=torch.float32)
if self.bits == 2:
self.gptqmodel_cuda.vecquant2matmul(
x.float(),
self.qweight,
out,
self.scales.float(),
self.qzeros,
self.g_idx,
)
elif self.bits == 3:
self.gptqmodel_cuda.vecquant3matmul(
x.float(),
self.qweight,
out,
self.scales.float(),
self.qzeros,
self.g_idx,
)
elif self.bits == 4:
self.gptqmodel_cuda.vecquant4matmul(
x.float(),
self.qweight,
out,
self.scales.float(),
self.qzeros,
self.g_idx,
)
elif self.bits == 8:
self.gptqmodel_cuda.vecquant8matmul(
x.float(),
self.qweight,
out,
self.scales.float(),
self.qzeros,
self.g_idx,
)
else:
if self.wf.device != self.qzeros.device:
self.wf = self.wf.to(self.qzeros.device)

if self.bits in [2, 4, 8]:
zeros = torch.bitwise_right_shift(
torch.unsqueeze(self.qzeros, 2).expand(-1, -1, 32 // self.bits),
self.wf.unsqueeze(0),
).to(torch.int16 if self.bits == 8 else torch.int8)
zeros = torch.bitwise_and(zeros, (2**self.bits) - 1)
if x.device.type != "cuda":
raise NotImplementedError(f"Unable to use cuda kernel. x.device.type is {x.device.type}")

zeros = zeros.reshape(self.scales.shape)

weight = torch.bitwise_right_shift(
torch.unsqueeze(self.qweight, 1).expand(-1, 32 // self.bits, -1),
self.wf.unsqueeze(-1),
).to(torch.int16 if self.bits == 8 else torch.int8)
weight = torch.bitwise_and(weight, (2**self.bits) - 1)
elif self.bits == 3:
zeros = self.qzeros.reshape(self.qzeros.shape[0], self.qzeros.shape[1] // 3, 3, 1).expand(
-1, -1, -1, 12
)
zeros = zeros >> self.wf.unsqueeze(0)
zeros[:, :, 0, 10] = (zeros[:, :, 0, 10] & 0x3) | ((zeros[:, :, 1, 0] << 2) & 0x4)
zeros[:, :, 1, 11] = (zeros[:, :, 1, 11] & 0x1) | ((zeros[:, :, 2, 0] << 1) & 0x6)
zeros = zeros & 0x7
zeros = torch.cat(
[zeros[:, :, 0, :11], zeros[:, :, 1, 1:12], zeros[:, :, 2, 1:11]],
dim=2,
)

zeros = zeros.reshape(self.scales.shape)
if not self.gptqmodel_cuda_available:
raise ValueError(
f"Trying to use the cuda backend, but could not import the C++/CUDA dependencies with the following error: {cuda_import_exception}"
)

weight = self.qweight.reshape(self.qweight.shape[0] // 3, 3, 1, self.qweight.shape[1]).expand(
-1, -1, 12, -1
)
weight = (weight >> self.wf.unsqueeze(-1)) & 0x7
weight[:, 0, 10] = (weight[:, 0, 10] & 0x3) | ((weight[:, 1, 0] << 2) & 0x4)
weight[:, 1, 11] = (weight[:, 1, 11] & 0x1) | ((weight[:, 2, 0] << 1) & 0x6)
weight = weight & 0x7
weight = torch.cat([weight[:, 0, :11], weight[:, 1, 1:12], weight[:, 2, 1:11]], dim=1)
if self.kernel_switch_threshold != 0 and x.shape[0] >= self.kernel_switch_threshold:
raise ValueError(
f"Trying to use the cuda backend, x.shape[0] is {x.shape[0]}, x.shape[0] cannot be greater than kernel_switch_threshold{self.kernel_switch_threshold}"
)

weight = weight.reshape(weight.shape[0] * weight.shape[1], weight.shape[2])
num_itr = self.g_idx.shape[0] // x.shape[-1]
if num_itr == 1:
weights = self.scales[self.g_idx.long()] * (weight - zeros[self.g_idx.long()])
else:
num_dim = self.g_idx.shape[0] // num_itr
weights = []
for i in range(num_itr):
scale_i = self.scales[:, i * num_dim : (i + 1) * num_dim]
weight_i = weight[:, i * num_dim : (i + 1) * num_dim]
zeros_i = zeros[:, i * num_dim : (i + 1) * num_dim]
g_idx_i = self.g_idx[i * num_dim : (i + 1) * num_dim]
weights.append(scale_i[g_idx_i.long()] * (weight_i - zeros_i[g_idx_i.long()]))
weights = torch.cat(weights, dim=1)
out = torch.matmul(x, weights)
out = torch.zeros((x.shape[0], self.outfeatures), device=x.device, dtype=torch.float32)
if self.bits == 2:
self.gptqmodel_cuda.vecquant2matmul(
x.float(),
self.qweight,
out,
self.scales.float(),
self.qzeros,
self.g_idx,
)
elif self.bits == 3:
self.gptqmodel_cuda.vecquant3matmul(
x.float(),
self.qweight,
out,
self.scales.float(),
self.qzeros,
self.g_idx,
)
elif self.bits == 4:
self.gptqmodel_cuda.vecquant4matmul(
x.float(),
self.qweight,
out,
self.scales.float(),
self.qzeros,
self.g_idx,
)
elif self.bits == 8:
self.gptqmodel_cuda.vecquant8matmul(
x.float(),
self.qweight,
out,
self.scales.float(),
self.qzeros,
self.g_idx,
)
out = out.to(x_dtype)
out = out.reshape(out_shape)
out = out + self.bias if self.bias is not None else out
Expand Down
Loading

0 comments on commit c0f8b11

Please sign in to comment.