[luci/pass] Add fp32_to_uint8_cast to header #14296
Conversation
01000-you
changed the title
01000-you
changed the title
|
plz check #13480 (comment) |
| @@ -292,11 +292,9 @@ uint32_t cal_offset(loco::TensorShape &dimension, uint32_t *indices) | |||
| indices[2] * dimension.dim(3).value() + indices[3]; | |||
| } | |||
|
|
|||
| uint32_t cal_offset_2d(loco::TensorShape &dimension, uint32_t *indices) | |||
| uint32_t cal_offset_2d(loco::TensorShape &dimension, uint32_t indices[2]) | |||
There was a problem hiding this comment.
This is odd.
I don't like adding 2d suffix and don't like adding [2] too.
There was a problem hiding this comment.
I understand that it's not easy to add it like that.
I can use the original functioncal_offset. I will remove it
There was a problem hiding this comment.
I don't think caller code would come up with a nice code.
It would be better to revise existing cal_offset to accept current 4d and new 2d.
There was a problem hiding this comment.
And it would be better to split these two.
They don't have common context.
01000-you
force-pushed
the
quantutils
branch
3 times, most recently
from
d5740bb
to
3bd0fa2
Compare
|
Current change and commit title/description doesn't match. |
01000-you
changed the title
|
Current first commit message is
|
01000-you
force-pushed
the
quantutils
branch
2 times, most recently
from
3f3d46a
to
2193541
Compare
Fixed it, thank you |
|
Your commit message doesn't match with the change. |
01000-you
force-pushed
the
quantutils
branch
2 times, most recently
from
2e6da7c
to
abdc9c5
Compare
|
can you plz add
|
This commit add the fp32_to_uint8_cast in the header file. ONE-DCO-1.0-Signed-off-by: y01000.you <[email protected]>
01000-you
force-pushed
the
quantutils
branch
from
abdc9c5
to
cf57989
Compare
seanshpark
removed
the
PR/NO MERGE
This commit add the
fp32_to_uint8_castin the header file.Related Issue : #13480
Draft PR: #13585
ONE-DCO-1.0-Signed-off-by: y01000.you [email protected]