def _scalar_dequant_v100_v2(x_l0c, deq_ub, align_shape, x_shape, relu_flag,
sqrt_mode):
"""
dequant for scale in v100
"""
res = tvm.compute(
align_shape,
lambda i, j, k, l:
(x_l0c(i, j, k, l).astype("float16") * deq_ub(0, 0, 0, 0)),
name='dequant_to_fp16')
if sqrt_mode:
res = tvm.compute(x_shape,
lambda i, j, k, l:
(res(i, j, k, l) * deq_ub(0, 0, 0, 0)),
name='dequant_sqrt')
if relu_flag:
res = tvm.compute(x_shape,
lambda *indices: tvm.relu(res(*indices)),
name="dequant_relu")
res = tvm.compute(x_shape,
lambda *indice: res(*indice),
name="res",
tag='dequant_res',
attrs={
'sqrt_mode': sqrt_mode,
'relu_mode': relu_flag,
'is_scalar': 1
})
return res
def _scalar_dequant_v100(x, x_shape, align_shape, deq_scale, relu_flag,
sqrt_mode):
"""
dequant for scale in v100
"""
res_f16 = tvm.compute(
align_shape,
lambda i, j, k, l:
(x(i, j, k, l).astype("float16") * deq_scale(0, 0, 0, 0, 0)),
name='dequant1',
tag="dequant1_scale")
res = tvm.compute(x_shape,
lambda *indice: res_f16(*indice),
name='dequant_remove_pad',
tag="dequant_remove_pad")
if relu_flag:
res = tvm.compute(x_shape,
lambda *indices: tvm.relu(res(*indices)),
name="dequant_relu",
tag="dequant_relu")
if sqrt_mode:
res = tvm.compute(
x_shape,
lambda i, j, k, l: (res(i, j, k, l) * deq_scale(0, 0, 0, 0, 0)),
name='dequant2',
tag='dequant2_scale',
)
return res
def _s16_to_s8_normal_compute(x, x1, req_scale, x_shape, align_shape, c1_index,
tensor_flag, relu_flag):
"""
generate s16_to_s8 compute
"""
if x1 is not None:
if relu_flag:
res_s16 = tvm.compute(
x_shape,
lambda *indices: tvm.relu(x(*indices) + x1(*indices)),
name="res_s16",
tag="requant_s16_vaddrelu")
else:
res_s16 = tvm.compute(x_shape,
lambda *indices: x(*indices) + x1(*indices),
name="res_s16",
tag="requant_s16_vadd")
else:
if relu_flag:
res_s16 = tvm.compute(x_shape,
lambda *indices: tvm.relu(x(*indices)),
name="res_s16",
tag="requant_s16_relu")
else:
res_s16 = tvm.compute(x_shape,
lambda *indices: x(*indices),
name="res_s16",
tag="requant_s16")
x_shape_list = te.lang.cce.util.shape_to_list(x_shape)
if tensor_flag:
res_ub = tvm.compute(align_shape,
_deq_cast_compute(res_s16, req_scale, align_shape,
c1_index, tensor_flag,
x_shape_list),
name='s16_to_s8',
tag="requant_s16_vector")
else:
res_ub = tvm.compute(align_shape,
_deq_cast_compute(res_s16, req_scale, align_shape,
c1_index, tensor_flag,
x_shape_list),
name='s16_to_s8',
tag="requant_s16_scale")
return res_s16, res_ub
def _vector_depthwise_fused_v100(x, x_shape, align_shape, deq_scale, relu_flag,
sqrt_mode):
"""
dequant for vector in v100
"""
if relu_flag:
res_f16 = tvm.compute(align_shape,
lambda i, j, a, k, l: tvm.relu(
x(i, j // 2, j % 2, k, l).astype("float16") *
deq_scale(0, j, 0, 0, l)),
name='dequant1',
tag="dequant1_vector",
attrs={"relu_flag": 1})
else:
res_f16 = tvm.compute(align_shape,
lambda i, j, a, k, l: x(i, j // 2, j % 2, k, l).
astype("float16") * deq_scale(0, j, a, 0, l),
name='dequant1',
tag="dequant1_vector",
attrs={"relu_flag": 0})
align_shape[3] = x_shape[3].value
if not sqrt_mode:
res = tvm.compute(align_shape,
lambda *indice: res_f16(*indice),
name='dequant_remove_pad',
tag="dequant_remove_pad",
attrs={"sqrt_flag": 0})
else:
res_sqrt = tvm.compute(
align_shape,
lambda i, j, a, k, l:
(res_f16(i, j, a, k, l) * deq_scale(0, j, a, 0, l)),
name='dequant2',
tag='dequant2_vector')
res = tvm.compute(align_shape,
lambda *indice: res_sqrt(*indice),
name='dequant2_remove_pad',
tag="dequant2_remove_pad",
attrs={"sqrt_flag": 1})
return res
def _vector_dequant_v100(x, x_shape, align_shape, deq_scale, relu_flag,
sqrt_mode):
"""
dequant for vector in v100
"""
if relu_flag:
res_f16 = tvm.compute(
align_shape,
lambda i, j, k, l: tvm.relu(
x(i, j, k, l).astype("float16") * deq_scale(0, j, 0, 0, l)),
name='dequant1',
tag="dequant1_vector",
attrs={"relu_flag": 1})
else:
res_f16 = tvm.compute(align_shape,
lambda i, j, k, l: x(i, j, k, l).astype(
"float16") * deq_scale(0, j, 0, 0, l),
name='dequant1',
tag="dequant1_vector",
attrs={"relu_flag": 0})
res = tvm.compute(x_shape,
lambda *indice: res_f16(*indice),
name='dequant_remove_pad',
tag="dequant_remove_pad")
if sqrt_mode:
res = tvm.compute(x_shape,
lambda i, j, k, l:
(res(i, j, k, l) * deq_scale(0, j, 0, 0, l)),
name='dequant2',
tag='dequant2_vector')
return res
def _matmul_compute(x, x_shape, deq_scale, sqrt_mode, relu_flag,
shape_matmul_origin, c1_index, tensor_flag):
"""
dequant for matmul
"""
if _is_support_v200_instruction():
if tensor_flag:
res_f16 = tvm.compute(x_shape,
_matmul_vdeq_cast_compute(
x, deq_scale, x_shape, c1_index,
tensor_flag, relu_flag, True),
name='dequant',
tag="dequant_vector")
else:
res_f16 = tvm.compute(x_shape,
_matmul_vdeq_cast_compute(
x, deq_scale, x_shape, c1_index,
tensor_flag, relu_flag, True),
name='dequant',
tag="dequant_scale")
else:
if tensor_flag:
res_f16 = tvm.compute(
x_shape,
_matmul_vdeq_cast_compute(x, deq_scale, x_shape, c1_index,
tensor_flag, relu_flag, False),
name='dequant',
tag="dequant_vector",
)
else:
res_f16 = tvm.compute(
x_shape,
_matmul_vdeq_cast_compute(x, deq_scale, x_shape, c1_index,
tensor_flag, relu_flag, False),
name='dequant',
tag="dequant",
)
if sqrt_mode:
if tensor_flag:
res_f16 = tvm.compute(x_shape,
_matmul_vdeq_cast_compute(
res_f16, deq_scale, x_shape,
c1_index, tensor_flag, relu_flag,
False),
name='dequant_sqrt',
tag="dequant_vector_sqrt")
else:
res_f16 = tvm.compute(x_shape,
_matmul_vdeq_cast_compute(
res_f16, deq_scale, x_shape,
c1_index, tensor_flag, relu_flag,
False),
name='dequant_sqrt',
tag="dequant_sqrt")
if relu_flag:
res_f16 = tvm.compute(x_shape,
lambda *indices: tvm.relu(res_f16[indices]),
name="dequant_relu",
tag="dequant_relu")
if not _is_nz_format(x):
# convert fractal_z to ND
res_out = tvm.compute(
shape_matmul_origin,
lambda i, j: res_f16[j // 16, i // 16, i % 16, j % 16],
name='dequant_ND',
tag='dequant_ND',
attrs={'format': 'NC1HWC0'})
else:
# nz format
res_out = tvm.compute(x_shape,
lambda *i: res_f16[i],
name='dequant_NZ',
tag='dequant_NZ',
attrs={'format': 'FRACTAL_NZ'})
return res_out