Add flowers dataset for image classification model

python/paddle/v2/dataset/tests/flowers_test.py

@@ -0,0 +1,51 @@
+# Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import paddle.v2.dataset.flowers
+import unittest
+
+
+class TestFlowers(unittest.TestCase):
+    def check_reader(self, reader):
+        sum = 0
+        label = 0
+        size = 224 * 224 * 3
+        for l in reader():
+            self.assertEqual(l[0].size, size)
+            if l[1] > label:
+                label = l[1]
+            sum += 1
+        return sum, label
+
+    def test_train(self):
+        instances, max_label_value = self.check_reader(
+            paddle.v2.dataset.flowers.train())
+        self.assertEqual(instances, 1020)
+        self.assertEqual(max_label_value, 102)
+
+    def test_test(self):
+        instances, max_label_value = self.check_reader(
+            paddle.v2.dataset.flowers.test())
+        self.assertEqual(instances, 6149)
+        self.assertEqual(max_label_value, 102)
+
+    def test_valid(self):
+        instances, max_label_value = self.check_reader(
+            paddle.v2.dataset.flowers.valid())
+        self.assertEqual(instances, 1020)
+        self.assertEqual(max_label_value, 102)
+
+
+if __name__ == '__main__':
+    unittest.main()

python/paddle/v2/image.py

@@ -1,14 +1,14 @@
 import numpy as np
 try:
     import cv2
-except:
-    print(
-        "import cv2 error, please install opencv-python: pip install opencv-python"
-    )
+except ImportError:
+    cv2 = None
+
+from cv2 import resize
 
 __all__ = [
-    "load_image", "resize_short", "to_chw", "center_crop", "random_crop",
-    "left_right_flip", "simple_transform", "load_and_transform"
+    "load_image_bytes", "load_image", "resize_short", "to_chw", "center_crop",
+    "random_crop", "left_right_flip", "simple_transform", "load_and_transform"
 ]
 """
 This file contains some common interfaces for image preprocess.
@@ -28,6 +28,28 @@ the image layout as follows.
 """
 
 
+def load_image_bytes(bytes, is_color=True):
+    """
+    Load an color or gray image from bytes array.
+
+    Example usage:
+    
+    .. code-block:: python
+        with open('cat.jpg') as f:
+            im = load_image(f.read())
+
+    :param bytes: the input image bytes array.
+    :type file: str
+    :param is_color: If set is_color True, it will load and
+                     return a color image. Otherwise, it will
+                     load and return a gray image.
+    """
+    flag = 1 if is_color else 0
+    file_bytes = np.asarray(bytearray(bytes), dtype=np.uint8)
+    img = cv2.imdecode(file_bytes, flag)
+    return img
+
+
 def load_image(file, is_color=True):
     """
     Load an color or gray image from the file path.
@@ -76,7 +98,7 @@ def resize_short(im, size):
         h_new = size * h / w
     else:
         w_new = size * w / h
-    im = cv2.resize(im, (h_new, w_new), interpolation=cv2.INTER_CUBIC)
+    im = resize(im, (h_new, w_new), interpolation=cv2.INTER_CUBIC)
     return im
 
 

python/paddle/v2/reader/decorator.py

@@ -14,13 +14,15 @@
 
 __all__ = [
     'map_readers', 'buffered', 'compose', 'chain', 'shuffle',
-    'ComposeNotAligned', 'firstn'
+    'ComposeNotAligned', 'firstn', 'xmap'
 ]
 
 import itertools
 import random
 from Queue import Queue
 from threading import Thread
+from multiprocessing import Queue as MQueue
+from multiprocessing import Process
 
 
 def map_readers(func, *readers):
@@ -224,3 +226,74 @@ def firstn(reader, n):
             yield item
 
     return firstn_reader
+
+
+class XmapEndSignal():
+    pass
+
+
+def xmap(mapper, reader, process_num, buffer_size):
+    """
+    Use multiprocess to map samples from reader by a mapper defined by user.
+    And this function contains a buffered decorator.
+    :param mapper:  a function to map sample.
+    :type mapper: callable
+    :param reader: the data reader to read from
+    :type reader: callable
+    :param process_num: process number to handle original sample 
+    :type process_num: int
+    :param buffer_size: max buffer size
+    :type buffer_size: int
+    :return: the decarated reader
+    :rtype: callable
+    """
+    end = XmapEndSignal()
+    in_queue = MQueue(buffer_size)
+    out_queue = MQueue(buffer_size)
+
+    # define a worker to read samples from reader to in_queue
+    def read_worker(reader, in_queue):
+        for i in reader():
+            in_queue.put(i)
+        in_queue.put(end)
+
+    # start a read worker in a thread
+    t = Thread(target=read_worker, args=(reader, in_queue))
+    t.daemon = True
+    t.start()
+
+    # define a worker to handle samples from in_queue by mapper
+    # and put mapped samples into out_queue
+    def handle_worker(in_queue, out_queue, mapper):
+        sample = in_queue.get()
+        while not isinstance(sample, XmapEndSignal):
+            r = mapper(sample)
+            out_queue.put(r)
+            sample = in_queue.get()
+        in_queue.put(end)
+        out_queue.put(end)
+
+    # start several handle_workers
+    workers = []
+    for i in xrange(process_num):
+        worker = Process(
+            target=handle_worker, args=(in_queue, out_queue, mapper))
+        worker.daemon = True
+        workers.append(worker)
+    for w in workers:
+        w.start()
+
+    def xreader():
+        sample = out_queue.get()
+        while not isinstance(sample, XmapEndSignal):
+            yield sample
+            sample = out_queue.get()
+        finish = 1
+        while finish < process_num:
+            sample = out_queue.get()
+            if isinstance(sample, XmapEndSignal):
+                finish += 1
+            else:
+                yield sample
+
+    return xreader