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

RandomGamma should return an empty Tensor for empty alpha.

Browse Source 
Before this change, it returns an error "Input alpha should have non-zero element count" if empty `alpha` is presented with non-empty `shape`.

PiperOrigin-RevId: 239255074
This commit is contained in:
Brian Patton 2019-03-19 13:20:21 -07:00 committed by TensorFlower Gardener
parent 9369d376e7
commit a24cfc8d09
2 changed files with 5 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
tensorflow/core/kernels/random_op.cc
View File

@ -173,7 +173,7 @@ class RandomGammaOp : public OpKernel {
Tensor* samples_t = nullptr;
OP_REQUIRES_OK(ctx, ctx->allocate_output(0, samples_shape, &samples_t));
if (num_samples == 0) return;
if (samples_shape.num_elements() == 0) return;
using random::PhiloxRandom;

4
tensorflow/python/kernel_tests/random/random_gamma_test.py
View File

@ -53,6 +53,10 @@ class RandomGammaTest(test.TestCase):
return func
def testEmptySamplingNoError(self):
self.evaluate(random_ops.random_gamma(
[5], alpha=np.ones([2, 0, 3]), beta=np.ones([3]), dtype=dtypes.float32))
@test_util.run_deprecated_v1
def testMomentsFloat32(self):
self._testMoments(dtypes.float32)