Skip to content

Instantly share code, notes, and snippets.

@dlibenzi
Created June 10, 2020 20:52
Show Gist options
  • Save dlibenzi/6bc350d6747a229a75793d8b28fa4bfb to your computer and use it in GitHub Desktop.
Save dlibenzi/6bc350d6747a229a75793d8b28fa4bfb to your computer and use it in GitHub Desktop.
>>>>>> This is the difference (GOOD - XLA)
Difference Tensor:
(1,1,.,.) =
0.0000 0.0000 0.0000 -0.9593 -0.3904 0.0000 0.0000 0.0000
-0.9408 0.0000 -0.9346 -1.7807 0.0000 -0.5677 0.0000 0.0000
0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 -2.8547 0.0000
-0.8860 -0.5832 0.0000 -0.8090 -1.1559 -0.9040 0.0000 0.0000
0.0000 0.0000 0.0000 -2.8392 -0.7890 0.0000 0.0000 -0.5895
0.0000 0.0000 0.0000 0.0000 0.0000 -2.7308 -1.9320 0.0000
0.0000 0.0000 -0.8913 0.0000 0.0000 0.0000 0.0000 0.0000
(1,2,.,.) =
0.0000 -0.3958 -0.9147 0.0000 0.0000 0.0000 -0.9497 0.0000
-0.9811 0.0000 0.0000 0.0000 0.0000 0.0000 -1.3581 -0.9155
0.0000 0.0000 0.0000 0.0000 -0.8035 0.0000 0.0000 0.0000
0.0000 -1.9573 0.0000 0.0000 -1.5961 -1.6798 0.0000 0.0000
-0.9578 -0.6626 0.0000 0.0000 0.0000 -1.8747 0.0000 0.0000
0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 -2.9994 0.0000
0.0000 0.0000 0.0000 -0.6671 -1.7346 0.0000 0.0000 0.0000
0.0000 -0.3139 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
[ CPUFloatType{1,2,8,8} ]
>>>>> This is GOOD result
Compared Tensors:
(1,1,.,.) =
0.0000 0.9150 0.0000 0.9593 0.3904 0.0000 0.0000 0.0000
0.9408 0.1332 0.9346 0.5936 0.8694 0.5677 0.7411 0.0000
0.8854 0.0000 0.0000 0.0000 0.0000 0.0000 0.2969 0.8317
0.0000 0.2695 0.0000 0.0000 0.0000 0.0000 0.9516 0.0000
0.8860 0.5832 0.0000 0.8090 0.5779 0.9040 0.5547 0.0000
0.0000 0.3644 0.7104 0.9464 0.7890 0.0000 0.7886 0.5895
0.7539 0.1952 0.0000 0.3068 0.1165 0.9103 0.6440 0.0000
0.0000 0.4913 0.8913 0.0000 0.0000 0.0000 0.0000 0.0000
(1,2,.,.) =
0.0000 0.3958 0.9147 0.0000 0.0000 0.0000 0.9497 0.6666
0.9811 0.0874 0.0000 0.1088 0.1637 0.7025 0.6790 0.9155
0.0000 0.0000 0.0000 0.0000 0.8035 0.3813 0.7860 0.0000
0.0000 0.6524 0.6057 0.0000 0.7980 0.8399 0.0000 0.0000
0.9578 0.3313 0.0000 0.0000 0.0000 0.6249 0.4340 0.0000
0.5117 0.0000 0.0000 0.2247 0.0000 0.0000 0.9998 0.5944
0.6541 0.0337 0.0000 0.3336 0.5782 0.0000 0.2846 0.2007
0.0000 0.3139 0.4654 0.0000 0.0000 0.0000 0.0000 0.0000
[ CPUFloatType{1,2,8,8} ]
-vs-
>>>>>> This is XLA result
(1,1,.,.) =
0.0000 0.9150 0.0000 1.9186 0.7809 0.0000 0.0000 0.0000
1.8815 0.1332 1.8692 2.3743 0.8694 1.1354 0.7411 0.0000
0.8854 0.0000 0.0000 0.0000 0.0000 0.0000 0.2969 0.8317
0.0000 0.2695 0.0000 0.0000 0.0000 0.0000 3.8062 0.0000
1.7720 1.1664 0.0000 1.6179 1.7338 1.8080 0.5547 0.0000
0.0000 0.3644 0.7104 3.7856 1.5781 0.0000 0.7886 1.1789
0.7539 0.1952 0.0000 0.3068 0.1165 3.6411 2.5761 0.0000
0.0000 0.4913 1.7826 0.0000 0.0000 0.0000 0.0000 0.0000
(1,2,.,.) =
0.0000 0.7917 1.8294 0.0000 0.0000 0.0000 1.8994 0.6666
1.9623 0.0874 0.0000 0.1088 0.1637 0.7025 2.0371 1.8309
0.0000 0.0000 0.0000 0.0000 1.6069 0.3813 0.7860 0.0000
0.0000 2.6098 0.6057 0.0000 2.3941 2.5197 0.0000 0.0000
1.9157 0.9939 0.0000 0.0000 0.0000 2.4996 0.4340 0.0000
0.5117 0.0000 0.0000 0.2247 0.0000 0.0000 3.9992 0.5944
0.6541 0.0337 0.0000 1.0007 2.3127 0.0000 0.2846 0.2007
0.0000 0.6279 0.4654 0.0000 0.0000 0.0000 0.0000 0.0000
[ CPUFloatType{1,2,8,8} ]
/home/dlibenzi/pytorch/xla/test/cpp/cpp_test_util.h:51: Failure
Value of: r
Actual: false
Expected: true
stride=1 pad=1 ceil=1 dilation=2
>>>>>>> This is the unpool input
INPUT
(1,1,.,.) =
0.8823 0.9150 0.3829 0.9593 0.3904 0.6009 0.2566 0.7936
0.9408 0.1332 0.9346 0.5936 0.8694 0.5677 0.7411 0.4294
0.8854 0.5739 0.2666 0.6274 0.2696 0.4414 0.2969 0.8317
0.1053 0.2695 0.3588 0.1994 0.5472 0.0062 0.9516 0.0753
0.8860 0.5832 0.3376 0.8090 0.5779 0.9040 0.5547 0.3423
0.6343 0.3644 0.7104 0.9464 0.7890 0.2814 0.7886 0.5895
0.7539 0.1952 0.0050 0.3068 0.1165 0.9103 0.6440 0.7071
0.6581 0.4913 0.8913 0.1447 0.5315 0.1587 0.6542 0.3278
(1,2,.,.) =
0.6532 0.3958 0.9147 0.2036 0.2018 0.2018 0.9497 0.6666
0.9811 0.0874 0.0041 0.1088 0.1637 0.7025 0.6790 0.9155
0.2418 0.1591 0.7653 0.2979 0.8035 0.3813 0.7860 0.1115
0.2477 0.6524 0.6057 0.3725 0.7980 0.8399 0.1374 0.2331
0.9578 0.3313 0.3227 0.0162 0.2137 0.6249 0.4340 0.1371
0.5117 0.1585 0.0758 0.2247 0.0624 0.1816 0.9998 0.5944
0.6541 0.0337 0.1716 0.3336 0.5782 0.0600 0.2846 0.2007
0.5014 0.3139 0.4654 0.1612 0.1568 0.2083 0.3289 0.1054
[ CPUFloatType{1,2,8,8} ]
>>>>>>> These are the indices
INDICES
(1,1,.,.) =
9 8 11 10 11 12 13 14
1 16 3 4 3 4 23 22
25 8 11 10 11 30 13 30
33 32 35 36 37 36 37 38
41 42 43 44 43 30 47 30
33 32 35 36 53 54 53 54
57 58 43 58 43 44 47 46
49 48 51 52 53 54 53 54
(1,2,.,.) =
9 8 11 12 13 14 15 14
1 2 1 2 21 6 7 6
25 8 25 28 29 28 15 14
33 32 33 20 37 20 37 22
25 26 25 28 29 46 29 46
33 32 51 52 37 52 37 38
57 40 57 58 43 46 47 46
49 48 51 52 51 52 55 54
[ CPULongType{1,2,8,8} ]
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment