library(tfdatasets)
dataset <- tensor_slices_dataset(c(1,2,3,4,5)) %>%
dataset_interleave(cycle_length = 2, block_length = 4, function(x) {
tensors_dataset(x) %>%
dataset_repeat(6)
})
# resulting dataset (newlines indicate "block" boundaries):
c(1, 1, 1, 1,
2, 2, 2, 2,
1, 1,
2, 2,
3, 3, 3, 3,
4, 4, 4, 4,
3, 3,
4, 4,
5, 5, 5, 5,
5, 5,
) dataset_interleave
Maps map_func across this dataset, and interleaves the results
Description
Maps map_func across this dataset, and interleaves the results
Usage
dataset_interleave(dataset, map_func, cycle_length, block_length = 1) Arguments
| Arguments | Description |
|---|---|
| dataset | A dataset |
| map_func | A function mapping a nested structure of tensors (having shapes and types defined by output_shapes() and output_types() to a dataset. |
| cycle_length | The number of elements from this dataset that will be processed concurrently. |
| block_length | The number of consecutive elements to produce from each input element before cycling to another input element. |
Details
The cycle_length and block_length arguments control the order in which elements are produced. cycle_length controls the number of input elements that are processed concurrently. In general, this transformation will apply map_func to cycle_length input elements, open iterators on the returned dataset objects, and cycle through them producing block_length consecutive elements from each iterator, and consuming the next input element each time it reaches the end of an iterator.
Examples
See Also
Other dataset methods: dataset_batch(), dataset_cache(), dataset_collect(), dataset_concatenate(), dataset_decode_delim(), dataset_filter(), dataset_map_and_batch(), dataset_map(), dataset_padded_batch(), dataset_prefetch_to_device(), dataset_prefetch(), dataset_reduce(), dataset_repeat(), dataset_shuffle_and_repeat(), dataset_shuffle(), dataset_skip(), dataset_take_while(), dataset_take(), dataset_window()