experiments/STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Allow sparse tensor with reorder work on potentially large dims,

Browse Source 
Update: review comments addressed.

Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
This commit is contained in:
Yong Tang 2020-12-17 00:00:10 +00:00
parent f137ea2a23
commit 63690f568c
2 changed files with 6 additions and 19 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
tensorflow/core/kernels/sparse_reorder_op.cc
View File

@ -27,7 +27,6 @@ limitations under the License.
#include "tensorflow/core/framework/types.h" #include "tensorflow/core/framework/types.h"
#include "tensorflow/core/lib/gtl/inlined_vector.h" #include "tensorflow/core/lib/gtl/inlined_vector.h"
#include "tensorflow/core/util/sparse/sparse_tensor.h" #include "tensorflow/core/util/sparse/sparse_tensor.h"
#include "tensorflow/core/util/overflow.h"
namespace tensorflow { namespace tensorflow {
@ -55,21 +54,10 @@ class SparseReorderOp : public OpKernel {
"Input shape should be a vector but received shape ", "Input shape should be a vector but received shape ",
input_shape_in.shape().DebugString())); input_shape_in.shape().DebugString()));
// Check if the sparse tensor input shape is valid gtl::ArraySlice<int64> input_shape(
int64 total = 1; input_shape_in.vec<int64>().data(), input_shape_in.NumElements());
for (int64 i = 0; i < input_shape_in.NumElements(); ++i) {
int dim = input_shape_in.vec<int64>()(i);
OP_REQUIRES(context, (dim >= 0),
errors::InvalidArgument("Dimension ", dim, " must be >= 0"));
total = MultiplyWithoutOverflow(total, dim);
OP_REQUIRES(context, (total > 0),
errors::InvalidArgument(
"Shape would have more than 2**63 - 1 elements"));
}
const TensorShape input_shape(input_shape_in.vec<int64>()); gtl::InlinedVector<int64, 8> std_order(input_shape.size());
gtl::InlinedVector<int64, 8> std_order(input_shape.dims());
std::iota(std_order.begin(), std_order.end(), 0); std::iota(std_order.begin(), std_order.end(), 0);
// Check if the sparse tensor is already ordered correctly // Check if the sparse tensor is already ordered correctly

7
tensorflow/python/kernel_tests/sparse_reorder_op_test.py
View File

@ -21,7 +21,6 @@ from __future__ import print_function
import numpy as np import numpy as np
from tensorflow.python.framework import dtypes from tensorflow.python.framework import dtypes
from tensorflow.python.framework import errors_impl
from tensorflow.python.framework import sparse_tensor from tensorflow.python.framework import sparse_tensor
from tensorflow.python.framework import test_util from tensorflow.python.framework import test_util
from tensorflow.python.ops import array_ops from tensorflow.python.ops import array_ops
@ -132,9 +131,9 @@ class SparseReorderTest(test.TestCase):
dense_shape=[4096, 4096, 4096, 4096, 4096, 4096]) dense_shape=[4096, 4096, 4096, 4096, 4096, 4096])
self.assertAllEqual( self.assertAllEqual(
(4096, 4096, 4096, 4096, 4096, 4096), sp_input.get_shape()) (4096, 4096, 4096, 4096, 4096, 4096), sp_input.get_shape())
with self.assertRaisesRegex(errors_impl.InvalidArgumentError, sp_output = sparse_ops.sparse_reorder(sp_input)
"Shape would have more than"): self.assertAllEqual(
sp_output = sparse_ops.sparse_reorder(sp_input) (4096, 4096, 4096, 4096, 4096, 4096), sp_output.get_shape())
if __name__ == "__main__": if __name__ == "__main__":