dingpeifei
87e41aaeee
IR operators of GPU and CPU are unified as batchnorm
2021-03-18 19:02:28 +08:00
He Wei
7d9a783993
[auto-monad] Support side-effects by auto-monad
...
The basic idea is: exploits data dependency to control the execution order
of side-effect operations, and keep the semantics of ANF unchanged.
The ControlDepend primitive is removed and there are two primitives added:
1. UpdateState:
```
a = Assign(para, value)
```
became:
```
a = Assign(para, value, u)
u = UpdateState(u, a)
```
2. Load:
```
x = Add(para, value)
```
became:
```
p = Load(para, u)
x = Add(p, value)
u = UpdateState(u, p)
```
2021-02-08 09:01:15 +08:00
l00591931
9ec100d069
Change TensorAdd to Add, from r1.1 to master
2021-02-01 17:53:52 +08:00
b51b3a6764
update Pool's attr kernel_size, pad_mode
2021-01-18 12:10:20 +08:00
chenzomi
ca85fcef40
[ME] change some format code.
2020-10-21 12:14:54 +08:00
chenzomi
d471d32e87
[ME] change `check_integer` to format `check_positive_int` and `check_integeter`
2020-10-10 12:08:10 +08:00
Wei Luning
24a10225cf
change base class of ref to tensor in cpp
2020-08-26 12:41:25 +08:00
huangdongrun
f30418991c
refactor bool op parsing to be consistent with pynative mode
...
add testcase of st
2020-08-20 16:45:30 +08:00
huangdongrun
2a6d346d2f
support if by if grad parameter
...
add join for ref
adjust env eliminate to eliminate all env ops
add partial app cache
resolve while endless
fix env eliminate
support for "for while" cases
fix join shape error
2020-08-15 09:47:10 +08:00
2c4cb49a11
support interface 'all' and 'any' of tensor
2020-08-04 18:24:56 +08:00
kpy
570da089a8
set output value for dynamic graph
2020-07-23 17:05:47 +08:00
f4cb445ea8
syn code for 0715
2020-07-16 17:18:21 +08:00
jiangjinsheng
7f9bbfd338
add Conv1d ops
2020-07-14 18:59:45 +08:00
4e832b2309
support implicit type conversion for pynative mode
2020-07-09 15:41:41 +08:00
66bbdb4a31
change tensor dtype and shape from function to attr
2020-06-12 19:03:23 +08:00
38436f929f
move hook function to primtivePy class
2020-06-10 15:33:46 +08:00
BowenK
96379faa3a
Remove ZerosLikeTensor and sub with ZerosLike
2020-06-02 09:52:12 +08:00
rick_sanchez
e2a322b6b7
Modify code to support dynamic graph.
2020-06-01 10:20:24 +08:00
86d197dfeb
clean pylint
2020-05-29 23:00:00 +08:00
fbdba6e4da
clean pylint
2020-05-22 13:23:13 +08:00
5a914994ba
clean pylint
2020-05-18 16:42:35 +08:00
26fd75895d
pylint waring clean
2020-05-13 11:30:27 +08:00
68bdcb4e62
!663 fix tensor equal bug
...
Merge pull request !663 from flywind/fix_bug
2020-04-27 21:23:20 +08:00
kpy
e64c755ad6
change tensor equal bug
2020-04-27 17:50:17 +08:00
2078229436
Add prim name to error message for array_ops
2020-04-26 15:23:04 +08:00
b812b18c02
support update parameter for vm
2020-04-21 20:19:46 +08:00
7c233a57fa
support python func print and != for list with none
2020-04-20 19:55:50 +08:00
75fec82b52
resolve pynative operator issue
2020-04-16 21:02:22 +08:00
2122a28e1a
fix bug in tests of vm conv2d
2020-04-15 17:44:19 +08:00
94589ce611
!226 expend conv stride and dilation to 2d
...
Merge pull request !226 from wangnan39/expend_conv_stride_to_2d
2020-04-14 14:58:22 +08:00
2604acedcb
extend conv stride and dilation to 2d
2020-04-14 13:16:27 +08:00
5841fe010e
Support pow's second input could be tensor and fix bug in bprop of pow
2020-04-11 16:16:58 +08:00
7541d3b067
Develop op MaxPoolWithArgMax
2020-04-08 20:47:49 +08:00
930a1fb0a8
initial version
...
Signed-off-by: leonwanghui <leon.wanghui@huawei.com>
2020-03-27 22:54:54 +08:00
dingpeifei
He Wei
l00591931
yuchaojie
chenzomi
Wei Luning
huangdongrun
kpy
jiangjinsheng
BowenK
rick_sanchez