PaddlePaddle · heavengate · Nov 11, 2020 · Nov 4, 2020
diff --git a/python/paddle/__init__.py b/python/paddle/__init__.py
@@ -27,8 +27,6 @@
      import paddle from the source directory; please install paddlepaddle*.whl firstly.'''
                      )
 
-import paddle.reader
-import paddle.dataset
 import paddle.batch
 batch = batch.batch
 from .fluid import monkey_patch_variable

diff --git a/python/paddle/dataset/__init__.py b/python/paddle/dataset/__init__.py
@@ -29,18 +29,5 @@
 import paddle.dataset.voc2012
 import paddle.dataset.image
 
-__all__ = [
-    'mnist',
-    'imikolov',
-    'imdb',
-    'cifar',
-    'movielens',
-    'conll05',
-    'uci_housing',
-    'wmt14',
-    'wmt16',
-    'mq2007',
-    'flowers',
-    'voc2012',
-    'image',
-]
+# set __all__ as empty for not showing APIs under paddle.dataset
+__all__ = []
diff --git a/python/paddle/dataset/cifar.py b/python/paddle/dataset/cifar.py
@@ -32,6 +32,7 @@
 import itertools
 import numpy
 import paddle.dataset.common
+import paddle.utils.deprecated as deprecated
 import tarfile
 import six
 from six.moves import cPickle as pickle
@@ -75,6 +76,10 @@ def reader():
     return reader
 
 
+@deprecated(
+    since="2.0.0",
+    update_to="paddle.vision.datasets.Cifar100",
+    reason="Please use new dataset API which supports paddle.io.DataLoader")
 def train100():
     """
     CIFAR-100 training set creator.
@@ -90,6 +95,10 @@ def train100():
         'train')
 
 
+@deprecated(
+    since="2.0.0",
+    update_to="paddle.vision.datasets.Cifar100",
+    reason="Please use new dataset API which supports paddle.io.DataLoader")
 def test100():
     """
     CIFAR-100 test set creator.
@@ -105,6 +114,10 @@ def test100():
         'test')
 
 
+@deprecated(
+    since="2.0.0",
+    update_to="paddle.vision.datasets.Cifar10",
+    reason="Please use new dataset API which supports paddle.io.DataLoader")
 def train10(cycle=False):
     """
     CIFAR-10 training set creator.
@@ -123,6 +136,10 @@ def train10(cycle=False):
         cycle=cycle)
 
 
+@deprecated(
+    since="2.0.0",
+    update_to="paddle.vision.datasets.Cifar10",
+    reason="Please use new dataset API which supports paddle.io.DataLoader")
 def test10(cycle=False):
     """
     CIFAR-10 test set creator.
@@ -141,6 +158,10 @@ def test10(cycle=False):
         cycle=cycle)
 
 
+@deprecated(
+    since="2.0.0",
+    update_to="paddle.vision.datasets.Cifar10",
+    reason="Please use new dataset API which supports paddle.io.DataLoader")
 def fetch():
     paddle.dataset.common.download(CIFAR10_URL, 'cifar', CIFAR10_MD5)
     paddle.dataset.common.download(CIFAR100_URL, 'cifar', CIFAR100_MD5)
diff --git a/python/paddle/dataset/conll05.py b/python/paddle/dataset/conll05.py
@@ -27,6 +27,7 @@
 import itertools
 import paddle.dataset.common
 import paddle.compat as cpt
+import paddle.utils.deprecated as deprecated
 from six.moves import zip, range
 
 __all__ = ['test, get_dict', 'get_embedding']
@@ -202,6 +203,10 @@ def reader():
     return reader
 
 
+@deprecated(
+    since="2.0.0",
+    update_to="paddle.text.datasets.Conll05st",
+    reason="Please use new dataset API which supports paddle.io.DataLoader")
 def get_dict():
     """
     Get the word, verb and label dictionary of Wikipedia corpus.
@@ -215,13 +220,21 @@ def get_dict():
     return word_dict, verb_dict, label_dict
 
 
+@deprecated(
+    since="2.0.0",
+    update_to="paddle.text.datasets.Conll05st",
+    reason="Please use new dataset API which supports paddle.io.DataLoader")
 def get_embedding():
     """
     Get the trained word vector based on Wikipedia corpus.
     """
     return paddle.dataset.common.download(EMB_URL, 'conll05st', EMB_MD5)
 
 
+@deprecated(
+    since="2.0.0",
+    update_to="paddle.text.datasets.Conll05st",
+    reason="Please use new dataset API which supports paddle.io.DataLoader")
 def test():
     """
     Conll05 test set creator.
@@ -242,6 +255,10 @@ def test():
     return reader_creator(reader, word_dict, verb_dict, label_dict)
 
 
+@deprecated(
+    since="2.0.0",
+    update_to="paddle.text.datasets.Conll05st",
+    reason="Please use new dataset API which supports paddle.io.DataLoader")
 def fetch():
     paddle.dataset.common.download(WORDDICT_URL, 'conll05st', WORDDICT_MD5)
     paddle.dataset.common.download(VERBDICT_URL, 'conll05st', VERBDICT_MD5)

diff --git a/python/paddle/dataset/flowers.py b/python/paddle/dataset/flowers.py
@@ -39,6 +39,7 @@
 from paddle.dataset.image import *
 from paddle.reader import map_readers, xmap_readers
 from paddle import compat as cpt
+import paddle.utils.deprecated as deprecated
 import os
 import numpy as np
 from multiprocessing import cpu_count
@@ -143,6 +144,10 @@ def reader():
         return map_readers(mapper, reader)
 
 
+@deprecated(
+    since="2.0.0",
+    update_to="paddle.vision.datasets.Flowers",
+    reason="Please use new dataset API which supports paddle.io.DataLoader")
 def train(mapper=train_mapper, buffered_size=1024, use_xmap=True, cycle=False):
     '''
     Create flowers training set reader.
@@ -172,6 +177,10 @@ def train(mapper=train_mapper, buffered_size=1024, use_xmap=True, cycle=False):
         cycle=cycle)
 
 
+@deprecated(
+    since="2.0.0",
+    update_to="paddle.vision.datasets.Flowers",
+    reason="Please use new dataset API which supports paddle.io.DataLoader")
 def test(mapper=test_mapper, buffered_size=1024, use_xmap=True, cycle=False):
     '''
     Create flowers test set reader.
@@ -201,6 +210,10 @@ def test(mapper=test_mapper, buffered_size=1024, use_xmap=True, cycle=False):
         cycle=cycle)
 
 
+@deprecated(
+    since="2.0.0",
+    update_to="paddle.vision.datasets.Flowers",
+    reason="Please use new dataset API which supports paddle.io.DataLoader")
 def valid(mapper=test_mapper, buffered_size=1024, use_xmap=True):
     '''
     Create flowers validation set reader.

diff --git a/python/paddle/dataset/imdb.py b/python/paddle/dataset/imdb.py
@@ -23,6 +23,7 @@
 from __future__ import print_function
 
 import paddle.dataset.common
+import paddle.utils.deprecated as deprecated
 import collections
 import tarfile
 import re
@@ -76,6 +77,10 @@ def build_dict(pattern, cutoff):
     return word_idx
 
 
+@deprecated(
+    since="2.0.0",
+    update_to="paddle.text.datasets.Imdb",
+    reason="Please use new dataset API which supports paddle.io.DataLoader")
 def reader_creator(pos_pattern, neg_pattern, word_idx):
     UNK = word_idx['<unk>']
     INS = []
@@ -94,6 +99,10 @@ def reader():
     return reader
 
 
+@deprecated(
+    since="2.0.0",
+    update_to="paddle.text.datasets.Imdb",
+    reason="Please use new dataset API which supports paddle.io.DataLoader")
 def train(word_idx):
     """
     IMDB training set creator.
@@ -111,6 +120,10 @@ def train(word_idx):
         re.compile("aclImdb/train/neg/.*\.txt$"), word_idx)
 
 
+@deprecated(
+    since="2.0.0",
+    update_to="paddle.text.datasets.Imdb",
+    reason="Please use new dataset API which supports paddle.io.DataLoader")
 def test(word_idx):
     """
     IMDB test set creator.
@@ -128,6 +141,10 @@ def test(word_idx):
         re.compile("aclImdb/test/neg/.*\.txt$"), word_idx)
 
 
+@deprecated(
+    since="2.0.0",
+    update_to="paddle.text.datasets.Imdb",
+    reason="Please use new dataset API which supports paddle.io.DataLoader")
 def word_dict():
     """
     Build a word dictionary from the corpus.
@@ -139,5 +156,9 @@ def word_dict():
         re.compile("aclImdb/((train)|(test))/((pos)|(neg))/.*\.txt$"), 150)
 
 
+@deprecated(
+    since="2.0.0",
+    update_to="paddle.text.datasets.Imdb",
+    reason="Please use new dataset API which supports paddle.io.DataLoader")
 def fetch():
     paddle.dataset.common.download(URL, 'imdb', MD5)
diff --git a/python/paddle/dataset/imikolov.py b/python/paddle/dataset/imikolov.py
@@ -22,6 +22,7 @@
 from __future__ import print_function
 
 import paddle.dataset.common
+import paddle.utils.deprecated as deprecated
 import collections
 import tarfile
 import six
@@ -111,6 +112,10 @@ def reader():
     return reader
 
 
+@deprecated(
+    since="2.0.0",
+    update_to="paddle.text.datasets.Imikolov",
+    reason="Please use new dataset API which supports paddle.io.DataLoader")
 def train(word_idx, n, data_type=DataType.NGRAM):
     """
     imikolov training set creator.
@@ -131,6 +136,10 @@ def train(word_idx, n, data_type=DataType.NGRAM):
                           data_type)
 
 
+@deprecated(
+    since="2.0.0",
+    update_to="paddle.text.datasets.Imikolov",
+    reason="Please use new dataset API which supports paddle.io.DataLoader")
 def test(word_idx, n, data_type=DataType.NGRAM):
     """
     imikolov test set creator.
@@ -151,5 +160,9 @@ def test(word_idx, n, data_type=DataType.NGRAM):
                           data_type)
 
 
+@deprecated(
+    since="2.0.0",
+    update_to="paddle.text.datasets.Imikolov",
+    reason="Please use new dataset API which supports paddle.io.DataLoader")
 def fetch():
     paddle.dataset.common.download(URL, "imikolov", MD5)
diff --git a/python/paddle/dataset/mnist.py b/python/paddle/dataset/mnist.py
@@ -21,6 +21,7 @@
 from __future__ import print_function
 
 import paddle.dataset.common
+import paddle.utils.deprecated as deprecated
 import gzip
 import numpy
 import struct
@@ -88,6 +89,10 @@ def reader():
     return reader
 
 
+@deprecated(
+    since="2.0.0",
+    update_to="paddle.vision.datasets.MNIST",
+    reason="Please use new dataset API which supports paddle.io.DataLoader")
 def train():
     """
     MNIST training set creator.
@@ -105,6 +110,10 @@ def train():
                                        TRAIN_LABEL_MD5), 100)
 
 
+@deprecated(
+    since="2.0.0",
+    update_to="paddle.vision.datasets.MNIST",
+    reason="Please use new dataset API which supports paddle.io.DataLoader")
 def test():
     """
     MNIST test set creator.
@@ -121,6 +130,10 @@ def test():
         100)
 
 
+@deprecated(
+    since="2.0.0",
+    update_to="paddle.vision.datasets.MNIST",
+    reason="Please use new dataset API which supports paddle.io.DataLoader")
 def fetch():
     paddle.dataset.common.download(TRAIN_IMAGE_URL, 'mnist', TRAIN_IMAGE_MD5)
     paddle.dataset.common.download(TRAIN_LABEL_URL, 'mnist', TRAIN_LABEL_MD5)