vec_cpadd

Description

Adds elements (odd and even) between adjacent pairs:

Prototype

vec_cpadd(mask, dst, src, repeat_times, dst_rep_stride, src_rep_stride)

Parameters

For details, see Parameters. dst has the same data type as src.

Atlas 200/300/500 Inference Product : Tensors of type float16

Atlas Training Series Product : Tensors of type float16

Returns

None

Applicability

Atlas 200/300/500 Inference Product

Atlas Training Series Product

Restrictions

For details, see Restrictions.

Example

This example describes only basic functions. For more details, see Example.

from tbe import tik
tik_instance = tik.Tik()
src_gm = tik_instance.Tensor("float16", (128,), name="src_gm", scope=tik.scope_gm)
dst_gm = tik_instance.Tensor("float16", (64,), name="dst_gm", scope=tik.scope_gm)
src_ub = tik_instance.Tensor("float16", (128,), name="src_ub", scope=tik.scope_ubuf)
dst_ub = tik_instance.Tensor("float16", (64,), name="dst_ub", scope=tik.scope_ubuf)
# Move the user input from the Global Memory to the Unified Buffer.
tik_instance.data_move(src_ub, src_gm, 0, 1, 8, 0, 0)
tik_instance.vec_cpadd(128, dst_ub, src_ub, 1, 4, 8)
# Move the compute result from the Unified Buffer to the Global Memory.
tik_instance.data_move(dst_gm, dst_ub, 0, 1, 4, 0, 0)

tik_instance.BuildCCE(kernel_name="vec_cpadd", inputs=[src_gm], outputs=[dst_gm])

Result example:

Input (src_gm):
[-1.899    3.43     6.39     2.61    -1.996    9.68     0.699    4.395
 -0.2384   5.39    -5.58    -0.592   -8.52    -1.713   -0.977    8.53
 -7.043    4.348   -2.475    4.16    -7.48    -3.48     0.7246   9.266
  7.566    2.225    0.07605 -8.59    -3.506    0.1312  -7.598   -7.668
 -8.76    -8.9      0.3625   7.906    9.26     2.357    6.71     0.5034
 -9.484   -2.215   -5.188    4.316   -9.96     7.254    5.855    3.146
 -4.613   -4.715   -4.637    7.99    -9.11     7.59    -2.705   -8.26
 -7.84     7.047    3.244    7.31     1.431   -1.55     6.465    1.723
  2.998    4.51    -7.445   -3.293    3.705   -4.11     3.783    0.5117
  1.835    6.62     0.6226   7.043    1.572   -8.58    -3.37    -5.836
  9.72    -6.047   -2.977    9.86    -9.55    -2.402   -4.71     3.793
  8.625   -0.3523   7.88    -1.134    7.465   -4.598   -7.055   -9.04
 -0.979   -0.6177  -6.355   -1.605   -8.92     1.049   -0.723    7.066
  8.984    9.74     9.49     6.133   -6.87    -7.582   -4.863    2.217
  8.46     9.84     5.785    4.992    8.68    -1.767    2.395    8.414
  9.375    2.877   -3.016   -4.137   -1.899    6.22    -7.426    4.99   ]

Output (dst_gm):
[  1.53      9.        7.684     5.094     5.152    -6.17    -10.234
   7.555    -2.695     1.686   -10.96      9.99      9.79     -8.516
  -3.375   -15.266   -17.66      8.266    11.62      7.215   -11.7
  -0.871    -2.707     9.       -9.33      3.352    -1.52    -10.96
  -0.793    10.555    -0.11914   8.19      7.508   -10.734    -0.4043
   4.297     8.45      7.664    -7.008    -9.2       3.672     6.883
 -11.95     -0.918     8.27      6.746     2.867   -16.1      -1.597
  -7.96     -7.875     6.344    18.72     15.625   -14.45     -2.646
  18.31     10.78      6.914    10.81     12.25     -7.152     4.32
  -2.438  ]