Comments (9)
So, how to fix it in TF 1.11 ?
from sbnet.
Unfortunately, I don't have the time to investigate. Just wanted to mention it.
Sent from my Galaxy S9+ using FastHub
from sbnet.
We have limited resources for supporting/testing various combinations of versions, packages, linux distributions etc. We currently do not support python 3.6, looks like the error happens with python 3.6?
We also currently support tensorflow 1.8 for this project and haven't upgraded to 1.11 yet.
from sbnet.
Thanks @andrei-pokrovsky for the feedback, as I said above, I created this issue just to let you know that it is not working with TF 1.11.0.
I believe it has nothing to do with Python 3.5/3.6, nonetheless it is worth trying ;)
from sbnet.
@andrei-pokrovsky I can give you a hint, it seems that there is an issue with TF 1.11.0.
from sbnet.
Compiled on TF 1.12, but failed a couple of tests. To compile, edit file: /usr/local/lib/python3.5/dist-packages/tensorflow/include/absl/strings/string_view.h
and remove ABSL_ASSERT
from line 496
Output from make test
:
cd ../benchmark && bash run_all_unittests.bash # unit tests
test_basic (reduce_mask_tests.ReduceMaskTests) ... ok
test_larger (reduce_mask_tests.ReduceMaskTests) ... 2019-05-05 20:24:07.045607: E tensorflow/core/common_runtime/executor.cc:623] Executor failed to create kernel. Invalid argument: Default MaxPoolingOp only supports NHWC on device type CPU
[[{{node MaxPool}} = MaxPool[T=DT_FLOAT, data_format="NCHW", ksize=[1, 1, 5, 5], padding="VALID", strides=[1, 1, 4, 4], _device="/job:localhost/replica:0/task:0/device:GPU:0"](MaxPool-0-TransposeNHWCToNCHW-LayoutOptimizer)]]
ok
test_session (reduce_mask_tests.ReduceMaskTests)
Use cached_session instead. ... ok
test_session (sparse_conv_tests.SparseConv2DCustomTests)
Use cached_session instead. ... ok
test_sparse_conv2d_with_mask_same (sparse_conv_tests.SparseConv2DCustomTests) ... ok
test_session (sparse_conv_tests.SparseConv2DTests)
Use cached_session instead. ... ok
test_sparse_conv2d_correctness (sparse_conv_tests.SparseConv2DTests) ... ok
test_sparse_conv2d_matmul_correctness (sparse_conv_tests.SparseConv2DTests) ... ok
test_sparse_conv2d_same (sparse_conv_tests.SparseConv2DTests) ... ok
test_sparse_conv2d_valid (sparse_conv_tests.SparseConv2DTests) ... ok
test_sparse_conv2d_with_large_block_strides (sparse_conv_tests.SparseConv2DTests) ... ok
test_sparse_conv2d_with_mask_same (sparse_conv_tests.SparseConv2DTests) ... ok
test_sparse_conv2d_with_mask_same_even_block (sparse_conv_tests.SparseConv2DTests) ... ok
test_sparse_conv2d_with_mask_same_even_block_strides (sparse_conv_tests.SparseConv2DTests) ... ok
test_sparse_conv2d_with_mask_valid (sparse_conv_tests.SparseConv2DTests) ... ok
test_offset_array (sparse_conv_tests.UpsampleIndicesTests) ... ok
test_session (sparse_conv_tests.UpsampleIndicesTests)
Use cached_session instead. ... ok
test_upsample_indices (sparse_conv_tests.UpsampleIndicesTests) ... ok
test_basic (sparse_gather_tests.SparseGatherTests) ... ok
test_large (sparse_gather_tests.SparseGatherTests) ... 2019-05-05 20:24:24.806640: E tensorflow/core/common_runtime/executor.cc:623] Executor failed to create kernel. Invalid argument: Default MaxPoolingOp only supports NHWC on device type CPU
[[{{node MaxPool}} = MaxPool[T=DT_FLOAT, data_format="NCHW", ksize=[1, 1, 5, 5], padding="VALID", strides=[1, 1, 4, 4], _device="/job:localhost/replica:0/task:0/device:GPU:0"](MaxPool-0-TransposeNHWCToNCHW-LayoutOptimizer)]]
FAIL
test_session (sparse_gather_tests.SparseGatherTests)
Use cached_session instead. ... ok
test_resblock_gradients (sparse_res_block_tests.ResBlockGradientTests) ... /usr/local/lib/python3.5/dist-packages/tensorflow/python/util/tf_inspect.py:75: DeprecationWarning: inspect.getargspec() is deprecated, use inspect.signature() instead
return _inspect.getargspec(target)
-------------------------------------------------------
Dense Residual
name grad angle abs err
x 21.745 0.653
sub3/conv3/Conv2D:0 0.020 0.000
sub3/relu3:0 0.000 0.000
sub3/bn3/FusedBatchNorm:0 0.000 0.000
sub2/conv2/Conv2D:0 2.916 0.236
sub2/relu2:0 3.595 0.169
sub2/bn2/FusedBatchNorm:0 3.987 0.169
sub1/conv1/Conv2D:0 3.129 0.397
sub1/relu1:0 3.126 0.307
sub1/bn1/FusedBatchNorm:0 3.330 0.307
sub1/bn1/gamma:0 17.718 0.645
sub1/bn1/beta:0 12.391 0.355
sub1/conv1/w:0 15.586 0.825
sub2/bn2/gamma:0 1.541 0.287
sub2/bn2/beta:0 41.559 0.501
sub2/conv2/w:0 14.427 0.620
sub3/bn3/gamma:0 0.000 0.573
sub3/bn3/beta:0 0.020 0.000
sub3/conv3/w:0 0.020 0.000
ok
test_session (sparse_res_block_tests.ResBlockGradientTests)
Use cached_session instead. ... ok
test_session (sparse_res_block_tests.SparseConv2DGradientTests)
Use cached_session instead. ... ok
test_sparse_conv2d_gradient (sparse_res_block_tests.SparseConv2DGradientTests) ...
-------------------------------------------------------
Sparse Conv Layer
name grad angle abs err
x 0.059 0.000
w 0.000 0.000
ok
test_session (sparse_res_block_tests.SparseResBlockGradientTests)
Use cached_session instead. ... ok
test_sparse_resblock_gradients (sparse_res_block_tests.SparseResBlockGradientTests) ...
-------------------------------------------------------
Sparse Residual
name grad angle abs err
x 1.587 0.930
SparseScatter:0 0.040 0.000
SparseGather:0 1.601 0.930
sub3/bn3/FusedBatchNorm:0 0.000 0.000
sub3/conv3/Conv2D:0 0.028 0.000
sub3/relu3:0 0.000 0.000
sub2/conv2/Conv2D:0 0.052 0.000
sub2/relu2:0 0.044 0.000
sub2/bn2/FusedBatchNorm:0 0.044 0.000
sub1/conv1/Conv2D:0 1.386 0.432
sub1/relu1:0 1.011 0.285
sub1/bn1/FusedBatchNorm:0 1.011 0.285
sub1/bn1/gamma:0 0.028 0.000
sub1/bn1/beta:0 0.000 0.000
sub1/conv1/w:0 0.000 0.000
sub2/bn2/gamma:0 1.610 0.000
sub2/bn2/beta:0 0.000 0.000
sub2/conv2/w:0 0.000 0.000
sub3/bn3/gamma:0 0.000 0.000
sub3/bn3/beta:0 0.020 0.000
sub3/conv3/w:0 0.020 0.000
-------------------------------------------------------
Sparse Residual
name grad angle abs err
x 1.587 0.930
SparseScatter:0 0.040 0.000
SparseGather:0 1.601 0.930
sub3/bn3/FusedBatchNorm:0 0.000 0.000
sub3/conv3/Conv2D:0 0.028 0.000
sub3/relu3:0 0.000 0.000
sub2/conv2/Conv2D:0 0.052 0.000
sub2/relu2:0 0.044 0.000
sub2/bn2/FusedBatchNorm:0 0.044 0.000
sub1/conv1/Conv2D:0 1.386 0.432
sub1/relu1:0 1.011 0.285
sub1/bn1/FusedBatchNorm:0 1.011 0.285
sub1/bn1/gamma:0 0.028 0.000
sub1/bn1/beta:0 0.000 0.000
sub1/conv1/w:0 0.000 0.000
sub2/bn2/gamma:0 1.610 0.000
sub2/bn2/beta:0 0.000 0.000
sub2/conv2/w:0 0.000 0.000
sub3/bn3/gamma:0 0.000 0.000
sub3/bn3/beta:0 0.020 0.000
sub3/conv3/w:0 0.020 0.000
ok
test_basic (sparse_scatter_tests.SparseScatterTests) ... ok
test_session (sparse_scatter_tests.SparseScatterTests)
Use cached_session instead. ... ok
test_calc_out_size (tf_conv_dims_tests.CalcOutSizeDeconvTests) ... ok
test_session (tf_conv_dims_tests.CalcOutSizeDeconvTests)
Use cached_session instead. ... ok
test_calc_out_size (tf_conv_dims_tests.CalcOutSizeTests) ... ok
test_session (tf_conv_dims_tests.CalcOutSizeTests)
Use cached_session instead. ... ok
test_calc_padding (tf_conv_dims_tests.CalcPaddingTests) ... ok
test_calc_padding_err_ksize_list (tf_conv_dims_tests.CalcPaddingTests) ... ok
test_calc_padding_err_strides_list (tf_conv_dims_tests.CalcPaddingTests) ... ok
test_calc_padding_err_strides_tensor (tf_conv_dims_tests.CalcPaddingTests) ... ERROR::tensorflow:assertion failed: [Expect first and last dimension of `strides` = 1.] [Condition x == y did not hold element-wise:] [x (stack:0) = ] [2 1] [y (Const_1:0) = ] [1 1]
[[node assert_equal/Assert/AssertGuard/Assert (defined at /workspace/sbnet/sbnet_tensorflow/benchmark/tf_conv_dims.py:67) = Assert[T=[DT_STRING, DT_STRING, DT_STRING, DT_INT64, DT_STRING, DT_INT64], summarize=3, _device="/job:localhost/replica:0/task:0/device:CPU:0"](assert_equal/Assert/AssertGuard/Assert/Switch/_11, assert_equal/Assert/AssertGuard/Assert/data_0, assert_equal/Assert/AssertGuard/Assert/data_1, assert_equal/Assert/AssertGuard/Assert/data_2, assert_equal/Assert/AssertGuard/Assert/Switch_1/_13, assert_equal/Assert/AssertGuard/Assert/data_4, assert_equal/Assert/AssertGuard/Assert/Switch_2/_15)]]
[[{{node assert_equal/Assert/AssertGuard/Assert/_18}} = _Recv[client_terminated=false, recv_device="/job:localhost/replica:0/task:0/device:GPU:0", send_device="/job:localhost/replica:0/task:0/device:CPU:0", send_device_incarnation=1, tensor_name="edge_55_assert_equal/Assert/AssertGuard/Assert", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:GPU:0"]()]]
Caused by op 'assert_equal/Assert/AssertGuard/Assert', defined at:
File "/usr/lib/python3.5/runpy.py", line 184, in _run_module_as_main
"__main__", mod_spec)
File "/usr/lib/python3.5/runpy.py", line 85, in _run_code
exec(code, run_globals)
File "/usr/lib/python3.5/unittest/__main__.py", line 18, in <module>
main(module=None)
File "/usr/lib/python3.5/unittest/main.py", line 94, in __init__
self.runTests()
File "/usr/lib/python3.5/unittest/main.py", line 255, in runTests
self.result = testRunner.run(self.test)
File "/usr/lib/python3.5/unittest/runner.py", line 176, in run
test(result)
File "/usr/lib/python3.5/unittest/suite.py", line 84, in __call__
return self.run(*args, **kwds)
File "/usr/lib/python3.5/unittest/suite.py", line 122, in run
test(result)
File "/usr/lib/python3.5/unittest/suite.py", line 84, in __call__
return self.run(*args, **kwds)
File "/usr/lib/python3.5/unittest/suite.py", line 122, in run
test(result)
File "/usr/lib/python3.5/unittest/suite.py", line 84, in __call__
return self.run(*args, **kwds)
File "/usr/lib/python3.5/unittest/suite.py", line 122, in run
test(result)
File "/usr/lib/python3.5/unittest/case.py", line 648, in __call__
return self.run(*args, **kwds)
File "/usr/lib/python3.5/unittest/case.py", line 600, in run
testMethod()
File "/workspace/sbnet/sbnet_tensorflow/benchmark/tf_conv_dims_tests.py", line 109, in test_calc_padding_err_strides_tensor
p = calc_padding_4d(tf.shape(x), [2, 3, 1, 1], tf.constant(np.array([2, 1, 1, 1])), 'SAME')
File "/workspace/sbnet/sbnet_tensorflow/benchmark/tf_conv_dims.py", line 106, in calc_padding_4d
strides = _check_strides(strides)
File "/workspace/sbnet/sbnet_tensorflow/benchmark/tf_conv_dims.py", line 67, in _check_strides
message='Expect first and last dimension of `strides` = 1.')
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/ops/check_ops.py", line 390, in assert_equal
return control_flow_ops.Assert(condition, data, summarize=summarize)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/util/tf_should_use.py", line 189, in wrapped
return _add_should_use_warning(fn(*args, **kwargs))
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/ops/control_flow_ops.py", line 167, in Assert
guarded_assert = cond(condition, no_op, true_assert, name="AssertGuard")
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/util/deprecation.py", line 488, in new_func
return func(*args, **kwargs)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/ops/control_flow_ops.py", line 2097, in cond
orig_res_f, res_f = context_f.BuildCondBranch(false_fn)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/ops/control_flow_ops.py", line 1930, in BuildCondBranch
original_result = fn()
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/ops/control_flow_ops.py", line 165, in true_assert
condition, data, summarize, name="Assert")
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/ops/gen_logging_ops.py", line 52, in _assert
name=name)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/op_def_library.py", line 787, in _apply_op_helper
op_def=op_def)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/util/deprecation.py", line 488, in new_func
return func(*args, **kwargs)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/ops.py", line 3274, in create_op
op_def=op_def)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/ops.py", line 1770, in __init__
self._traceback = tf_stack.extract_stack()
InvalidArgumentError (see above for traceback): assertion failed: [Expect first and last dimension of `strides` = 1.] [Condition x == y did not hold element-wise:] [x (stack:0) = ] [2 1] [y (Const_1:0) = ] [1 1]
[[node assert_equal/Assert/AssertGuard/Assert (defined at /workspace/sbnet/sbnet_tensorflow/benchmark/tf_conv_dims.py:67) = Assert[T=[DT_STRING, DT_STRING, DT_STRING, DT_INT64, DT_STRING, DT_INT64], summarize=3, _device="/job:localhost/replica:0/task:0/device:CPU:0"](assert_equal/Assert/AssertGuard/Assert/Switch/_11, assert_equal/Assert/AssertGuard/Assert/data_0, assert_equal/Assert/AssertGuard/Assert/data_1, assert_equal/Assert/AssertGuard/Assert/data_2, assert_equal/Assert/AssertGuard/Assert/Switch_1/_13, assert_equal/Assert/AssertGuard/Assert/data_4, assert_equal/Assert/AssertGuard/Assert/Switch_2/_15)]]
[[{{node assert_equal/Assert/AssertGuard/Assert/_18}} = _Recv[client_terminated=false, recv_device="/job:localhost/replica:0/task:0/device:GPU:0", send_device="/job:localhost/replica:0/task:0/device:CPU:0", send_device_incarnation=1, tensor_name="edge_55_assert_equal/Assert/AssertGuard/Assert", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:GPU:0"]()]]
ok
test_calc_padding_stride (tf_conv_dims_tests.CalcPaddingTests) ... ok
test_calc_padding_valid (tf_conv_dims_tests.CalcPaddingTests) ... ok
test_session (tf_conv_dims_tests.CalcPaddingTests)
Use cached_session instead. ... ok
======================================================================
FAIL: test_large (sparse_gather_tests.SparseGatherTests)
----------------------------------------------------------------------
Traceback (most recent call last):
File "/workspace/sbnet/sbnet_tensorflow/benchmark/sparse_gather_tests.py", line 122, in test_large
self._test_sparse_gather(mask, x, w, bsize, ksize, strides, padding)
File "/workspace/sbnet/sbnet_tensorflow/benchmark/sparse_gather_tests.py", line 79, in _test_sparse_gather
np.testing.assert_array_equal(set(l1), set(l2))
File "/usr/local/lib/python3.5/dist-packages/numpy/testing/_private/utils.py", line 896, in assert_array_equal
verbose=verbose, header='Arrays are not equal')
File "/usr/local/lib/python3.5/dist-packages/numpy/testing/_private/utils.py", line 819, in assert_array_compare
raise AssertionError(msg)
AssertionError:
Arrays are not equal
Mismatch: 100%
x: array({(0.0, 28.0, 29.0), (0.0, 44.0, 45.0), (6.0, 7.0, 0.0), (0.0, 1.0, 2.0), (24.0, 25.0, 26.0), (9.0, 10.0, 11.0), (18.0, 19.0, 20.0), (39.0, 0.0, 36.0), (52.0, 53.0, 54.0), (27.0, 28.0, 32.0), (12.0, 13.0, 14.0), (37.0, 38.0, 39.0), (3.0, 4.0, 8.0), (36.0, 4.0, 5.0), (61.0, 62.0, 63.0), (55.0, 0.0, 60.0), (21.0, 22.0, 23.0), (0.0, 0.0, 0.0), (30.0, 31.0, 0.0), (46.0, 47.0, 0.0), (33.0, 34.0, 35.0), (15.0, 0.0, 20.0), (12.0, 16.0, 17.0), (36.0, 37.0, 38.0)},
dtype=object)
y: array({(0.0, 60.0, 61.0), (0.0, 12.0, 13.0), (0.0, 36.0, 37.0), (0.0, 1.0, 2.0), (20.0, 21.0, 22.0), (24.0, 25.0, 26.0), (23.0, 0.0, 28.0), (62.0, 63.0, 0.0), (14.0, 15.0, 0.0), (5.0, 6.0, 7.0), (9.0, 10.0, 11.0), (18.0, 19.0, 20.0), (29.0, 30.0, 31.0), (27.0, 28.0, 32.0), (0.0, 0.0, 4.0), (3.0, 4.0, 8.0), (38.0, 39.0, 0.0), (36.0, 36.0, 37.0), (0.0, 0.0, 0.0), (44.0, 45.0, 46.0), (33.0, 34.0, 35.0), (47.0, 0.0, 52.0), (53.0, 54.0, 55.0), (12.0, 16.0, 17.0)},
dtype=object)
----------------------------------------------------------------------
Ran 40 tests in 330.915s
FAILED (failures=1)
Makefile:14: recipe for target 'test' failed
make: *** [test] Error 1
from sbnet.
Anyone successfully complied the codes with newer tf (e.g. 1.15) and passed the test?
from sbnet.
Anyone successfully complied the codes with newer tf (e.g. 1.15) and passed the test?
Finally it worked, my environment setup:
CUDA 9.0 + tensorflow 1.12 + several modifications in cuda_runtime_api.h
Also, I set D_GLIBCXX_USE_CXX11_ABI=1 even though I am using g++ 5.4
from sbnet.
Can you tell me what are the modifications you made cuda_runtime_api.h?
from sbnet.
Related Issues (20)
- auto testing error HOT 2
- Running with variable input size HOT 2
- Is the detection model used in the SBNet paper open? HOT 2
- make test fails HOT 4
- AttributeError: module 'f21c708d1ddc75dcce283dd13fe531f7' has no attribute 'sparse_gather' HOT 4
- No speed-up despite sparsity HOT 15
- Usage within C++ inference pipeline HOT 1
- Performance for sparse_scatter_var worse not always better than sparse_scatter HOT 3
- Unable to cast to tf.Variable after sparse_scatter operation HOT 1
- CXX11_ABI Flag HOT 2
- No gradient defined for SparseScatter_Var HOT 4
- Invalid configuration argument reduce mask
- make test error 1 HOT 2
- sparse_conv_lib doesn't do sparse_conv2d_custom correctly HOT 1
- Huge loss in accuracy using SBNet HOT 3
- sample.py not working
- is there a pytorch version? HOT 4
- Time cost increases HOT 1
- Is there any way to reduce computation time of sbnet_module.reduce_mask? HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from sbnet.