Пример #1
0
    def testSparseExpandDims(self):
        for rank in range(1, 4):
            # Create a dummy input. When rank=3, shape=[2, 4, 6].
            shape = np.arange(1, rank + 1) * 2
            before = np.arange(np.prod(shape)).reshape(shape)

            # Make entries sparse.
            before *= np.random.binomial(1, .2, before.shape)
            dense_shape = before.shape
            indices = np.array(np.where(before)).T
            values = before[before != 0]

            # Try every possible valid value of axis.
            for axis in range(-rank - 1, rank):
                expected_after = np.expand_dims(before, axis)

                for axis_as_tensor in [False, True]:
                    dense_shape_t = constant_op.constant(dense_shape,
                                                         dtype=dtypes.int64)
                    indices_t = constant_op.constant(indices)
                    values_t = constant_op.constant(values)
                    before_t = sparse_tensor.SparseTensor(
                        indices=indices_t,
                        values=values_t,
                        dense_shape=dense_shape_t)

                    if axis_as_tensor:
                        axis = constant_op.constant(axis)

                    s = sparse_ops.sparse_expand_dims(before_t, axis)
                    d = sparse_ops.sparse_to_dense(s.indices, s.dense_shape,
                                                   s.values)
                    self.assertAllEqual(self.evaluate(d), expected_after)
Пример #2
0
  def testSparseExpandDims(self):
    for rank in range(1, 4):
      # Create a dummy input. When rank=3, shape=[2, 4, 6].
      shape = np.arange(1, rank + 1) * 2
      before = np.arange(np.prod(shape)).reshape(shape)

      # Make entries sparse.
      before *= np.random.binomial(1, .2, before.shape)
      dense_shape = before.shape
      indices = np.array(np.where(before)).T
      values = before[before != 0]

      # Try every possible valid value of axis.
      for axis in range(-rank - 1, rank):
        expected_after = np.expand_dims(before, axis)

        for axis_as_tensor in [False, True]:
          dense_shape_t = constant_op.constant(dense_shape, dtype=dtypes.int64)
          indices_t = constant_op.constant(indices)
          values_t = constant_op.constant(values)
          before_t = sparse_tensor.SparseTensor(
              indices=indices_t, values=values_t, dense_shape=dense_shape_t)

          if axis_as_tensor:
            axis = constant_op.constant(axis)

          s = sparse_ops.sparse_expand_dims(before_t, axis)
          d = sparse_ops.sparse_to_dense(s.indices, s.dense_shape, s.values)
          self.assertAllEqual(self.evaluate(d), expected_after)
 def expand_dims(inputs, axis):
     if tf_utils.is_sparse(inputs):
         return sparse_ops.sparse_expand_dims(inputs, axis)
     else:
         return array_ops.expand_dims(inputs, axis)