You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@mxnet.apache.org by GitBox <gi...@apache.org> on 2020/10/03 20:48:28 UTC
[GitHub] [incubator-mxnet] szha opened a new issue #19286: infer shape error in hybridized block for zero-size input
szha opened a new issue #19286:
URL: https://github.com/apache/incubator-mxnet/issues/19286
## Description
For a hybridized HybridBlock, the new forward interface throws an error for zero-size input
### Error Message
```
MXNetError: MXNetError: Shape inconsistent, Provided = [2,4,0,128], inferred shape=[2,4,4,128]
```
<details>
<summary>Complete error</summary>
---> 17 model(np.zeros((2, 2, 4, 0, 128)))
~/mxnet/python/mxnet/util.py in _with_np_shape(*args, **kwargs)
296 def _with_np_shape(*args, **kwargs):
297 with np_shape(active=True):
--> 298 return func(*args, **kwargs)
299 return _with_np_shape
300 else:
~/mxnet/python/mxnet/util.py in _with_np_array(*args, **kwargs)
480 def _with_np_array(*args, **kwargs):
481 with np_array(active=True):
--> 482 return func(*args, **kwargs)
483 return _with_np_array
484 else:
~/mxnet/python/mxnet/gluon/block.py in __call__(self, x, *args)
1427
1428 with x.ctx:
-> 1429 return self._call_cached_op(x, *args)
1430
1431 def forward(self, x, *args):
~/mxnet/python/mxnet/util.py in _with_np_shape(*args, **kwargs)
296 def _with_np_shape(*args, **kwargs):
297 with np_shape(active=True):
--> 298 return func(*args, **kwargs)
299 return _with_np_shape
300 else:
~/mxnet/python/mxnet/util.py in _with_np_array(*args, **kwargs)
480 def _with_np_array(*args, **kwargs):
481 with np_array(active=True):
--> 482 return func(*args, **kwargs)
483 return _with_np_array
484 else:
~/mxnet/python/mxnet/gluon/block.py in _call_cached_op(self, *args)
1096 def _call_cached_op(self, *args):
1097 if self._cached_op is None:
-> 1098 self._build_cache(*args)
1099 assert self._cached_op, "Gluon failed to build the cache. " \
1100 "This should never happen. " \
~/mxnet/python/mxnet/util.py in _with_np_shape(*args, **kwargs)
296 def _with_np_shape(*args, **kwargs):
297 with np_shape(active=True):
--> 298 return func(*args, **kwargs)
299 return _with_np_shape
300 else:
~/mxnet/python/mxnet/util.py in _with_np_array(*args, **kwargs)
480 def _with_np_array(*args, **kwargs):
481 with np_array(active=True):
--> 482 return func(*args, **kwargs)
483 return _with_np_array
484 else:
~/mxnet/python/mxnet/gluon/block.py in _build_cache(self, *args)
993
994 def _build_cache(self, *args):
--> 995 data, out = self._get_graph(*args)
996 data_names = {data.name: i for i, data in enumerate(data)}
997 params = {p.var().name: p for p in self.collect_params().values()}
~/mxnet/python/mxnet/util.py in _with_np_shape(*args, **kwargs)
296 def _with_np_shape(*args, **kwargs):
297 with np_shape(active=True):
--> 298 return func(*args, **kwargs)
299 return _with_np_shape
300 else:
~/mxnet/python/mxnet/util.py in _with_np_array(*args, **kwargs)
480 def _with_np_array(*args, **kwargs):
481 with np_array(active=True):
--> 482 return func(*args, **kwargs)
483 return _with_np_array
484 else:
~/mxnet/python/mxnet/gluon/block.py in _get_graph(self, *args)
989 return self._get_graph_v1(*args)
990 else: # Gluon 2 based on deferred compute mode
--> 991 return self._get_graph_v2(*args)
992 return self._cached_graph
993
~/mxnet/python/mxnet/util.py in _with_np_shape(*args, **kwargs)
296 def _with_np_shape(*args, **kwargs):
297 with np_shape(active=True):
--> 298 return func(*args, **kwargs)
299 return _with_np_shape
300 else:
~/mxnet/python/mxnet/util.py in _with_np_array(*args, **kwargs)
480 def _with_np_array(*args, **kwargs):
481 with np_array(active=True):
--> 482 return func(*args, **kwargs)
483 return _with_np_array
484 else:
~/mxnet/python/mxnet/gluon/block.py in _get_graph_v2(self, *args)
978 args = _regroup(flatten_args, self._in_format)
979 with autograd.pause(), dc.context():
--> 980 out = super().__call__(*args)
981 flatten_out, self._out_format = _flatten(out, "output")
982 symbol_outputs = dc.get_symbol(flatten_out, sym_cls=type(symbol_inputs[0]))
~/mxnet/python/mxnet/gluon/block.py in __call__(self, *args)
709 hook(self, args)
710
--> 711 out = self.forward(*args)
712
713 for hook in self._forward_hooks.values():
~/mxnet/python/mxnet/util.py in _with_np_shape(*args, **kwargs)
296 def _with_np_shape(*args, **kwargs):
297 with np_shape(active=True):
--> 298 return func(*args, **kwargs)
299 return _with_np_shape
300 else:
~/mxnet/python/mxnet/util.py in _with_np_array(*args, **kwargs)
480 def _with_np_array(*args, **kwargs):
481 with np_array(active=True):
--> 482 return func(*args, **kwargs)
483 return _with_np_array
484 else:
<ipython-input-1-164fad6d3e22> in forward(self, x)
8 super().__init__()
9 def forward(self, x):
---> 10 return x[0]
11
12
~/mxnet/python/mxnet/numpy/multiarray.py in __getitem__(self, key)
743 'index {} is out of bounds for axis 0 with size {}'.format(
744 key, shape[0]))
--> 745 return self._at(key)
746 elif isinstance(key, py_slice):
747 # Unlike numpy/_symbol.py, calls MXNDArraySlice64 writable memory
~/mxnet/python/mxnet/ndarray/ndarray.py in _at(self, idx)
1423 else:
1424 check_call(_LIB.MXNDArrayAt(
-> 1425 self.handle, ctypes.c_uint32(idx), ctypes.byref(handle)))
1426 return self.__class__(handle=handle, writable=self.writable)
1427
~/mxnet/python/mxnet/base.py in check_call(ret)
244 """
245 if ret != 0:
--> 246 raise get_last_ffi_error()
247
248
MXNetError: MXNetError: Shape inconsistent, Provided = [2,4,0,128], inferred shape=[2,4,4,128]
</details>
## To Reproduce
Run the following script
```
from mxnet import np, npx, gluon, use_np
npx.set_np()
@use_np
class TestModel(gluon.HybridBlock):
def __init__(self):
super().__init__()
def forward(self, x):
return x[0]
model = TestModel()
model.initialize()
model.hybridize()
model(np.zeros((2, 2, 4, 0, 128)))
```
## Environment
<details>
<summary>Environment Information</summary>
```
----------Python Info----------
('Version :', '3.7.8')
('Compiler :', 'GCC 4.2.1 Compatible Apple LLVM 11.0.3 (clang-1103.0.29.20) (-macos10.15-objc-')
('Build :', ('default', 'Jun 5 2020 22:59:21'))
('Arch :', ('64bit', ''))
------------Pip Info-----------
No corresponding pip install for current python.
----------MXNet Info-----------
No MXNet installed.
----------System Info----------
('Platform :', 'Darwin-19.6.0-x86_64-i386-64bit')
('system :', 'Darwin')
('node :', 'a483e79ab3ab')
('release :', '19.6.0')
('version :', 'Darwin Kernel Version 19.6.0: Mon Aug 31 22:12:52 PDT 2020; root:xnu-6153.141.2~1/RELEASE_X86_64')
----------Hardware Info----------
('machine :', 'x86_64')
('processor :', 'i386')
machdep.cpu.brand_string: Intel(R) Core(TM) i7-8569U CPU @ 2.80GHz
machdep.cpu.features: FPU VME DE PSE TSC MSR PAE MCE CX8 APIC SEP MTRR PGE MCA CMOV PAT PSE36 CLFSH DS ACPI MMX FXSR SSE SSE2 SS HTT TM PBE SSE3 PCLMULQDQ DTES64 MON DSCPL VMX EST TM2 SSSE3 FMA CX16 TPR PDCM SSE4.1 SSE4.2 x2APIC MOVBE POPCNT AES PCID XSAVE OSXSAVE SEGLIM64 TSCTMR AVX1.0 RDRAND F16C
machdep.cpu.leaf7_features: RDWRFSGS TSC_THREAD_OFFSET SGX BMI1 AVX2 SMEP BMI2 ERMS INVPCID FPU_CSDS MPX RDSEED ADX SMAP CLFSOPT IPT MDCLEAR TSXFA IBRS STIBP L1DF SSBD
machdep.cpu.extfeatures: SYSCALL XD 1GBPAGE EM64T LAHF LZCNT PREFETCHW RDTSCP TSCI
----------Network Test----------
Setting timeout: 10
Timing for MXNet: https://github.com/apache/incubator-mxnet, DNS: 0.0040 sec, LOAD: 2.5430 sec.
Timing for PYPI: https://pypi.python.org/pypi/pip, DNS: 0.6464 sec, LOAD: 9.0609 sec.
Timing for FashionMNIST: https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/dataset/fashion-mnist/train-labels-idx1-ubyte.gz, DNS: 0.0006 sec, LOAD: 2.1972 sec.
Error open Conda: https://repo.continuum.io/pkgs/free/, HTTP Error 403: Forbidden, DNS finished in 0.000483989715576 sec.
Timing for Gluon Tutorial(en): http://gluon.mxnet.io, DNS: 0.0009 sec, LOAD: 2.8254 sec.
Error open Gluon Tutorial(cn): https://zh.gluon.ai, <urlopen error [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed (_ssl.c:727)>, DNS finished in 0.00119805335999 sec.
----------Environment----------
CC="/usr/local/opt/llvm/bin/clang"
CXX="/usr/local/opt/llvm/bin/clang++"
```
</details>
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@mxnet.apache.org
For additional commands, e-mail: issues-help@mxnet.apache.org
[GitHub] [incubator-mxnet] szha closed issue #19286: infer shape error in hybridized block for zero-size input
Posted by GitBox <gi...@apache.org>.
szha closed issue #19286:
URL: https://github.com/apache/incubator-mxnet/issues/19286
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@mxnet.apache.org
For additional commands, e-mail: issues-help@mxnet.apache.org
[GitHub] [incubator-mxnet] leezu commented on issue #19286: infer shape error in hybridized block for zero-size input
Posted by GitBox <gi...@apache.org>.
leezu commented on issue #19286:
URL: https://github.com/apache/incubator-mxnet/issues/19286#issuecomment-703816274
This bug is related to legacy (non-numpy) reshape operator. Reshape with `0` implies "copy this dimension from input".
We can see that all shapes are correct prior to infer_shape pass:
```
frame #6: 0x00007fff3bd1cb8b libmxnet.so`mxnet::imperative::SetShapeType(ctx=0x00007fffffff8f18, attrs=0x00007fffffff97a8, inputs=size=1, outputs=size=1, dispatch_mode=0x00007fffffff8f24) at imperative_utils.h:208:26
205 common::ConvertToNumpyShape(&in_shapes);
206 common::ConvertToNumpyShape(&out_shapes);
207 }
-> 208 const bool success = infershape[attrs.op](attrs, &in_shapes, &out_shapes);
209 if (!success) {
210 std::stringstream os;
211 os << "Operator " << attrs.op->name << " inferring shapes failed.\n";
(lldb) parray 5 in_shapes[0].data_heap_
(long *) $15 = 0x0000555555f45700 {
(long) [0] = 1
(long) [1] = 2
(long) [2] = 4
(long) [3] = 0
(long) [4] = 128
}
(lldb) p out_shapes (mxnet::ShapeVector) $16 = size=1 {
[0] = {
mxnet::Tuple<long> = {
ndim_ = 4
num_heap_allocated_ = 0
data_stack_ = ([0] = 2, [1] = 4, [2] = 0, [3] = 128)
data_heap_ = 0x0000000000000000
}
}
}
```
But after infer_shape, `0` is replaced by `4`.
```
frame #1: 0x00007fff4968c1be libmxnet.so`mxnet::op::ReshapeShape(attrs=0x00007fffffff97a8, in_attrs=0x00005555569cbb68, out_attrs=0x00005555569cbb80) at matrix_op-inl.h:235:3
232 << "Target: " << oshape
233 << "\nSource: " << dshape;
234 #endif
-> 235 SHAPE_ASSIGN_CHECK(*out_attrs, 0, oshape);
236 return ReverseReshapeInferShape(&(*in_attrs)[0], (*out_attrs)[0]);
237 }
238
(lldb) p oshape
(mxnet::TShape) $17 = {
mxnet::Tuple<long> = {
ndim_ = 4
num_heap_allocated_ = 0
data_stack_ = ([0] = 2, [1] = 4, [2] = 4, [3] = 128)
data_heap_ = 0x0000000000000000
}
}
```
https://github.com/apache/incubator-mxnet/blob/f732530c8e1f8dcc11134e935430e3793ddbf4c8/src/operator/tensor/matrix_op-inl.h#L194-L237
The root cause is that `MXNDArrayAt`, `MXNDArrayReshape` and `MXNDArraySlice` do not sufficiently distinguish between the numpy and non-numpy mode and always record legacy operators.
We need to update the recording step to record numpy / legacy operator based on if numpy / legacy mode is enabled:
https://github.com/apache/incubator-mxnet/blob/72eff9b66ecc683c3e7f9ad2c0ba69efa8dd423b/src/ndarray/ndarray.cc#L302-L308
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@mxnet.apache.org
For additional commands, e-mail: issues-help@mxnet.apache.org