tf.raw_ops.DynamicEnqueueTPUEmbeddingRaggedTensorBatch
tf.raw_ops.DynamicEnqueueTPUEmbeddingRaggedTensorBatch(
sample_splits,
embedding_indices,
aggregation_weights,
mode_override,
device_ordinal,
table_ids,
combiners=[],
max_sequence_lengths=[],
num_features=[],
name=None
)
| Args |
|---|
sample_splits | A list of at least 1 Tensor objects with the same type in: int32, int64. |
embedding_indices | A list with the same length as sample_splits of Tensor objects with the same type in: int32, int64. |
aggregation_weights | A list with the same length as sample_splits of Tensor objects with the same type in: float32, float64. |
mode_override | A Tensor of type string. |
device_ordinal | A Tensor of type int32. |
table_ids | A list of ints. |
combiners | An optional list of strings. Defaults to []. |
max_sequence_lengths | An optional list of ints. Defaults to []. |
num_features | An optional list of ints. Defaults to []. |
name | A name for the operation (optional). |
| Returns |
|---|
| The created Operation. |