p47068915
/
mindspore
forked from mindspore-Ecosystem/mindspore
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

!1586 dataset: fix some format problem in take and split 

Merge pull request !1586 from ms_yan/take_split_format
This commit is contained in:
mindspore-ci-bot 2020-05-28 19:09:07 +08:00 committed by Gitee
parent 6d056e0c0a 5e8db93618
commit fb78bb6ece
1 changed files with 12 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

23
mindspore/dataset/engine/datasets.py
View File

@ -560,9 +560,9 @@ class Dataset:
Note:
1. If count is greater than the number of element in dataset or equal to -1,
all the element in dataset will be taken.
all the element in dataset will be taken.
2. The order of using take and batch effects. If take before batch operation,
then taken given number of rows, otherwise take given number of batches.
then taken given number of rows, otherwise take given number of batches.
Args:
count (int, optional): Number of elements to be taken from the dataset (default=-1).
@ -590,7 +590,7 @@ class Dataset:
# here again
dataset_size = self.get_dataset_size()
if(dataset_size is None or dataset_size <= 0):
if dataset_size is None or dataset_size <= 0:
raise RuntimeError("dataset size unknown, unable to split.")
all_int = all(isinstance(item, int) for item in sizes)
@ -640,8 +640,8 @@ class Dataset:
Note:
1. Dataset cannot be sharded if split is going to be called.
2. It is strongly recommended to not shuffle the dataset, but use randomize=True instead.
Shuffling the dataset may not be deterministic, which means the data in each split
will be different in each epoch.
Shuffling the dataset may not be deterministic, which means the data in each split
will be different in each epoch.
Raises:
RuntimeError: If get_dataset_size returns None or is not supported for this dataset.
@ -1173,6 +1173,7 @@ class SourceDataset(Dataset):
def is_sharded(self):
raise NotImplementedError("SourceDataset must implement is_sharded.")
class MappableDataset(SourceDataset):
"""
Abstract class to represent a source dataset which supports use of samplers.
@ -1253,13 +1254,13 @@ class MappableDataset(SourceDataset):
Note:
1. Dataset should not be sharded if split is going to be called. Instead, create a
DistributedSampler and specify a split to shard after splitting. If dataset is
sharded after a split, it is strongly recommended to set the same seed in each instance
of execution, otherwise each shard may not be part of the same split (see Examples)
DistributedSampler and specify a split to shard after splitting. If dataset is
sharded after a split, it is strongly recommended to set the same seed in each instance
of execution, otherwise each shard may not be part of the same split (see Examples)
2. It is strongly recommended to not shuffle the dataset, but use randomize=True instead.
Shuffling the dataset may not be deterministic, which means the data in each split
will be different in each epoch. Furthermore, if sharding occurs after split, each
shard may not be part of the same split.
Shuffling the dataset may not be deterministic, which means the data in each split
will be different in each epoch. Furthermore, if sharding occurs after split, each
shard may not be part of the same split.
Raises:
RuntimeError: If get_dataset_size returns None or is not supported for this dataset.