| /external/tensorflow/tensorflow/python/ops/ragged/ |
| D | ragged_map_ops.py | 358 flat_values = rt_ta.flat_values.concat()
363 flat_values=flat_values,
375 flat_values = rt.flat_values
385 flat_values=flat_values,
396 values = t.flat_values
410 flat_values=d.dtype,
443 flat_values=current,
|
| D | ragged_tensor.py | 514 flat_values, argument
555 [flat_values] + list(nested_value_rowids) + list(nested_nrows)):
556 result = flat_values
563 def from_nested_row_splits(cls, flat_values, nested_row_splits, name=None): argument
586 [flat_values] + list(nested_row_splits)):
587 result = flat_values
593 def from_nested_row_lengths(cls, flat_values, nested_row_lengths, name=None): argument
616 [flat_values] + list(nested_row_lengths)):
617 result = flat_values
714 def flat_values(self): member in RaggedTensor
[all …]
|
| D | ragged_factory_ops.py | 284 flat_values = pylist
286 if not all(isinstance(v, (list, tuple)) for v in flat_values):
290 flat_values = sum((list(v) for v in flat_values), [])
294 inner_shape = get_inner_shape(flat_values)
295 check_inner_shape(flat_values, inner_shape)
|
| D | ragged_math_ops.py | 279 array_ops.ones_like(data.flat_values), data.nested_row_splits)
282 return total.with_flat_values(total.flat_values / count.flat_values)
293 array_ops.ones_like(data.flat_values), data.nested_row_splits)
297 total.flat_values / math_ops.sqrt(count.flat_values))
455 return reduce_op(rt_input.flat_values, None, name=name)
535 array_ops.ones_like(input_tensor.flat_values),
542 total.flat_values / count.flat_values, total.nested_row_splits)
|
| D | ragged_dispatch_test.py | 151 array_ops.shape(x.flat_values), array_ops.shape(y.flat_values))
238 dense_x = x.flat_values if isinstance(x, ragged_tensor.RaggedTensor) else x
246 result_flat_values = array_ops.reshape(result.flat_values, [-1])
360 dense_x = x.flat_values if isinstance(x, ragged_tensor.RaggedTensor) else x
361 dense_y = y.flat_values if isinstance(y, ragged_tensor.RaggedTensor) else y
370 result_flat_values = array_ops.reshape(result.flat_values, [-1])
425 x.flat_values if isinstance(x, ragged_tensor.RaggedTensor) else x
436 result_flat_values = array_ops.reshape(result.flat_values, [-1])
|
| D | ragged_string_ops.py | 75 if input_tensor.flat_values.shape.ndims > 1:
80 unicode_encode(input_tensor.flat_values, output_encoding, errors,
363 input.flat_values,
368 flat_input = array_ops.reshape(input.flat_values, [-1])
|
| D | ragged_dispatch.py | 132 flat_values = [
133 elt.flat_values if ragged_tensor.is_ragged(elt) else elt
139 self._original_op(flat_values, *args, **kwargs),
146 mapped_values = self._original_op(x.flat_values, *args, **kwargs)
201 (x_is_ragged and x.flat_values.shape.ndims <= y.shape.ndims) or
202 (y_is_ragged and y.flat_values.shape.ndims <= x.shape.ndims)):
211 x_values = x.flat_values if ragged_tensor.is_ragged(x) else x
212 y_values = y.flat_values if ragged_tensor.is_ragged(y) else y
|
| D | ragged_gather_ops.py | 112 params_dense_values=params.flat_values,
201 result = indices.with_flat_values(gather_nd(params, indices.flat_values))
215 params_ndims = params.ragged_rank + array_ops.rank(params.flat_values)
|
| D | ragged_concat_ops.py | 203 flat_values = [rt.flat_values for rt in rt_inputs]
204 concatenated_flat_values = array_ops.concat(flat_values, axis=0)
|
| D | ragged_tensor_shape.py | 172 array_ops.shape(rt_input.flat_values)[1:])
509 rt_input.flat_values.shape.ndims - 1 - dst_shape.num_inner_dimensions)
513 rt_input.flat_values, ragged_rank=inner_rank_diff))
536 rt_input.flat_values,
569 dst_values = ragged_util.repeat_ranges(rt_input.flat_values, splits,
|
| D | ragged_array_ops.py | 323 ragged_tiled_values = array_ops.gather(rt_input.flat_values, inner_value_ids)
512 return array_ops.size(input.flat_values, out_type=out_type, name=name)
578 return input.ragged_rank + array_ops.rank(input.flat_values)
|
| D | ragged_tensor_test.py | 156 flat_values=[3, 1, 4, 1, 5, 9, 2, 6],
188 self.assertAllEqual(values, rt_value.flat_values)
202 self.assertAllEqual(values, rt_value.flat_values)
522 flat_values = constant_op.constant(['a', 'b', 'c', 'd', 'e', 'f', 'g'])
528 rt = RaggedTensor.from_nested_row_splits(flat_values, nested_row_splits)
538 self.assertIs(rt_values_values, flat_values)
619 self.assertAllEqual(rt.flat_values,
648 self.eval_to_list(rt.flat_values),
681 self.eval_to_list(rt.flat_values),
1123 rt2_times_10 = rt2.with_flat_values(rt2.flat_values * 10)
|
| D | ragged_to_sparse_op_test.py | 193 [rt1.flat_values, rt2.flat_values])
|
| D | ragged_functional_ops.py | 103 return value.flat_values
|
| D | ragged_tensor_value.py | 62 def flat_values(self): member in RaggedTensorValue
|
| D | ragged_constant_value_op_test.py | 170 self.assertEqual(rt.flat_values.shape[1:], inner_shape)
|
| /external/tensorflow/tensorflow/contrib/framework/python/ops/ |
| D | script_ops.py | 127 flat_values = _py_func(
140 for ret_t, shape in zip(flat_values, flattened_shapes):
143 return nest.pack_sequence_as(output_types, flat_values)
|
| /external/tensorflow/tensorflow/tools/api/golden/v2/ |
| D | tensorflow.-ragged-tensor.pbtxt | 11 name: "flat_values"
48 …argspec: "args=[\'cls\', \'flat_values\', \'nested_row_lengths\', \'name\'], varargs=None, keyword…
52 …argspec: "args=[\'cls\', \'flat_values\', \'nested_row_splits\', \'name\'], varargs=None, keywords…
56 …argspec: "args=[\'cls\', \'flat_values\', \'nested_value_rowids\', \'nested_nrows\', \'name\'], va…
|
| /external/tensorflow/tensorflow/tools/api/golden/v1/ |
| D | tensorflow.-ragged-tensor.pbtxt | 11 name: "flat_values"
48 …argspec: "args=[\'cls\', \'flat_values\', \'nested_row_lengths\', \'name\'], varargs=None, keyword…
52 …argspec: "args=[\'cls\', \'flat_values\', \'nested_row_splits\', \'name\'], varargs=None, keywords…
56 …argspec: "args=[\'cls\', \'flat_values\', \'nested_value_rowids\', \'nested_nrows\', \'name\'], va…
|
| D | tensorflow.ragged.-ragged-tensor-value.pbtxt | 10 name: "flat_values"
|
| /external/tensorflow/tensorflow/core/api_def/base_api/ |
| D | api_def_RaggedGather.pbtxt | 14 The `flat_values` for the `params` RaggedTensor. There was a terminology change
15 at the python level from dense_values to flat_values, so dense_values is the
35 description: "The `flat_values` for the returned RaggedTensor."
|
| D | api_def_RaggedTensorToSparse.pbtxt | 10 description: "The `flat_values` for the `RaggedTensor`."
|
| D | api_def_RaggedRange.pbtxt | 22 description: "The `flat_values` for the returned `RaggedTensor`."
|
| /external/tensorflow/tensorflow/python/ops/ |
| D | ctc_ops.py | 861 flat_values = array_ops.reshape(dense, [-1])
863 array_ops.shape(flat_values, out_type=dtypes.int64)[0])
868 values = array_ops.boolean_mask(flat_values, flat_mask)
871 dense_shape=array_ops.shape(flat_values, out_type=dtypes.int64))
|
| /external/tensorflow/tensorflow/core/kernels/boosted_trees/ |
| D | quantile_ops.cc | 479 auto flat_values = values_tensor.flat<float>(); in Compute() local
487 const float value = flat_values(instance); in Compute()
|