You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by GitBox <gi...@apache.org> on 2022/08/05 08:40:08 UTC
[GitHub] [tvm] huanmei9 opened a new pull request, #12315: [TOPI][OP]change float multiplication of resize op to integer division
huanmei9 opened a new pull request, #12315:
URL: https://github.com/apache/tvm/pull/12315
This PR is aim to change floating point multiplication of resize operator to integer division when `method` is nearest_neighbor and `coordinate_transformation_mode` is asymmetric.
**Why this change?**
When using `create_prim_func` to convert topi.upsampling into primfunc, tir analyzer cannot parse complex floating point operations in tir block body. In the following case, the dimensions of read buffer are inferred as 0 : 128. We expect to be `tir.reads([x[i0_1, i1_1, i2_1/scale_h, i3_1/scale_w]])`in line 18，so it can perform compute_at schedule on this prim_func.
```python
scale_h = 2
scale_w = 2
x = te.placeholder([1, 128, 128, 128], "int8", "x")
y = topi.nn.upsampling(x, scale_h , scale_w)
func = te.create_prim_func([x, y])
print(func)
```
```Python
primfn(var_x: handle, var_resize: handle) -> ()
attr = {"global_symbol": "main", "tir.noalias": True}
buffers = {x: Buffer(x_1: Pointer(global int8), int8, [1, 128, 128, 128], []),
resize: Buffer(resize_1: Pointer(global int8), int8, [1, 128, 256, 256], [])}
buffer_map = {var_x: x, var_resize: resize} {
block([], "root") {
tir.reads([])
tir.writes([])
for (i0: int32, 0, 1) {
for (i1: int32, 0, 128) {
for (i2: int32, 0, 256) {
for (i3: int32, 0, 256) {
block([1, 128, 256, 256], "resize") as [i0_1, i1_1, i2_1, i3_1] {
bind(i0_1, i0)
bind(i1_1, i1)
bind(i2_1, i2)
bind(i3_1, i3)
tir.reads([x[i0_1, i1_1, 0:128, 0:128]])
tir.writes([resize[i0_1, i1_1, i2_1, i3_1]])
resize[i0_1, i1_1, i2_1, i3_1] = cast(int8, cast(float32, x[i0_1, i1_1, max(min(cast(int32, @tir.floor(((0.5f32*cast(float32, i2_1)) + 1e-05f32), dtype=float32)), 127), 0), max(min(cast(int32, @tir.floor(((0.5f32*cast(float32, i3_1)) + 1e-05f32), dtype=float32)), 127), 0)]))
}
}
}
}
}
```
**Why can it be changed?**
This PR only modify float multiplication to integer division when `method` is `nearest_neighbor` and `coordinate_transformation_mode` is `asymmetric`(it's default setting for topi.upsampling op).
Go further, we expect to convert $\lfloor s * x + \epsilon_1 \rfloor$ to $x//a$, $s$ is scale, a float number, $x$ is input, $\epsilon_1$ is 1e-5 in resize op, $a$ is the reciprocal of $s$, an integer.
So, we can represent output as this:
$y = \lfloor s * x + \epsilon_1 \rfloor = \lfloor ( \frac{1}{a} + \epsilon_0 ) * x + \epsilon_1 \rfloor = \lfloor \frac{x}{a} + \epsilon_0 * x + \epsilon_1 \rfloor$, let $x = a * m + n$, where $m = x // a, n = x \% a$ , then we can get $y = m + \lfloor \frac{n}{a} + \epsilon_0 * x + \epsilon_1 \rfloor$. In the case of upsampling op, $x$ and $a$ are always integers, so inequality $\frac{n}{a} \le \frac{a-1}{a}$ always holds. If $\epsilon_0 * x + \epsilon_1 < \frac{1}{a}$, we can get $0 < \frac{n}{a} + \epsilon_0 * x + \epsilon_1 < 1$，then $y = m + \lfloor \frac{n}{a} + \epsilon_0 * x + \epsilon_1 \rfloor = m = x // a$.
To sum up, when $\epsilon_0 * x + \epsilon_1 < \frac{1}{a}$, we can get $y = \lfloor s*x + \epsilon_1 \rfloor = x // a$.
In this PR, we assume $\epsilon_0 = \epsilon_1 = 1e-5$ and `can_multiple2div`to check $\epsilon_0 * x + \epsilon_1 < \frac{1}{a}$.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] Hzfengsy commented on pull request #12315: [TOPI][OP]change float multiplication of resize op to integer division
Posted by GitBox <gi...@apache.org>.
Hzfengsy commented on PR #12315:
URL: https://github.com/apache/tvm/pull/12315#issuecomment-1209257222
cc @masahi
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] wrongtest-intellif commented on a diff in pull request #12315: [TOPI][OP]change float multiplication of resize op to integer division
Posted by GitBox <gi...@apache.org>.
wrongtest-intellif commented on code in PR #12315:
URL: https://github.com/apache/tvm/pull/12315#discussion_r941005802
##########
tests/python/topi/python/test_topi_upsampling.py:
##########
@@ -101,6 +101,38 @@ def check_target(target, dev):
check_target(target, dev)
+@tvm.testing.uses_gpu
Review Comment:
why uses_gpu?
##########
python/tvm/topi/image/resize.py:
##########
@@ -23,6 +23,22 @@
from .. import tag
+def can_multiple2div(image, target):
Review Comment:
The names seem to be not so clear. Can they be named as below?
`can_multiple2div` -> `can_convert_multiply_to_intdiv`
`image` -> `origin_size`
`target` -> `scaled_size`
##########
tests/python/topi/python/test_topi_upsampling.py:
##########
@@ -101,6 +101,38 @@ def check_target(target, dev):
check_target(target, dev)
+@tvm.testing.uses_gpu
+def test_int_div_upsampling():
Review Comment:
can you kindly describe the purpose of this check in the docstring?
##########
python/tvm/topi/image/resize.py:
##########
@@ -23,6 +23,22 @@
from .. import tag
+def can_multiple2div(image, target):
+ """Check whether can transform multiplion to division"""
Review Comment:
typo `multiplion` -> `multiplication`?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] wrongtest-intellif merged pull request #12315: [TOPI][OP]change float multiplication of resize op to integer division
Posted by GitBox <gi...@apache.org>.
wrongtest-intellif merged PR #12315:
URL: https://github.com/apache/tvm/pull/12315
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] huanmei9 commented on pull request #12315: [TOPI][OP]change float multiplication of resize op to integer division
Posted by GitBox <gi...@apache.org>.
huanmei9 commented on PR #12315:
URL: https://github.com/apache/tvm/pull/12315#issuecomment-1210064900
@tvm-bot rerun
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] huanmei9 commented on a diff in pull request #12315: [TOPI][OP]change float multiplication of resize op to integer division
Posted by GitBox <gi...@apache.org>.
huanmei9 commented on code in PR #12315:
URL: https://github.com/apache/tvm/pull/12315#discussion_r941078556
##########
tests/python/topi/python/test_topi_upsampling.py:
##########
@@ -101,6 +101,38 @@ def check_target(target, dev):
check_target(target, dev)
+@tvm.testing.uses_gpu
Review Comment:
Thanks for your comment. No ideal about this setting, is requires_llvm better?
##########
tests/python/topi/python/test_topi_upsampling.py:
##########
@@ -101,6 +101,38 @@ def check_target(target, dev):
check_target(target, dev)
+@tvm.testing.uses_gpu
+def test_int_div_upsampling():
Review Comment:
Great suggestion, i will add it asap.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org