Merge pull request #290 from qingqing01/image

fix bug in 03.image_classification

03.image_classification/README.en.md

@@ -364,8 +364,7 @@ momentum_optimizer = paddle.optimizer.Momentum(
     learning_rate=0.1 / 128.0,
     learning_rate_decay_a=0.1,
     learning_rate_decay_b=50000 * 100,
-    learning_rate_schedule='discexp',
-    batch_size=128)
+    learning_rate_schedule='discexp')
 
 # Create trainer
 trainer = paddle.trainer.SGD(cost=cost,
@@ -483,7 +482,7 @@ Figure 12. The error rate of VGG model on CIFAR10
 
 ## Application
 
-After training is done, users can use the trained model to classify images. The following code shows how to infer through `paddle.infer` interface.
+After training is done, users can use the trained model to classify images. The following code shows how to infer through `paddle.infer` interface. You can remove the comments to change the model name.
 
 ```python
 from PIL import Image
@@ -492,13 +491,26 @@ import os
 def load_image(file):
     im = Image.open(file)
     im = im.resize((32, 32), Image.ANTIALIAS)
-    im = np.array(im).astype(np.float32).flatten()
+    im = np.array(im).astype(np.float32)
+    # The storage order of the loaded image is W(widht),
+    # H(height), C(channel). PaddlePaddle requires
+    # the CHW order, so transpose them.
+    im = im.transpose((2, 0, 1)) # CHW
+    # In the training phase, the channel order of CIFAR
+    # image is B(Blue), G(green), R(Red). But PIL open
+    # image in RGB mode. It must swap the channel order.
+    im = im[(2, 1, 0),:,:] # BGR
+    im = im.flatten()
     im = im / 255.0
     return im
 test_data = []
 cur_dir = os.path.dirname(os.path.realpath(__file__))
 test_data.append((load_image(cur_dir + '/image/dog.png'),)
 
+# users can remove the comments and change the model name
+# with gzip.open('params_pass_50.tar.gz', 'r') as f:
+#    parameters = paddle.parameters.Parameters.from_tar(f)
+
 probs = paddle.infer(
     output_layer=out, parameters=parameters, input=test_data)
 lab = np.argsort(-probs) # probs and lab are the results of one batch data

03.image_classification/README.md

@@ -356,8 +356,7 @@ momentum_optimizer = paddle.optimizer.Momentum(
     learning_rate=0.1 / 128.0,
     learning_rate_decay_a=0.1,
     learning_rate_decay_b=50000 * 100,
-    learning_rate_schedule='discexp',
-    batch_size=128)
+    learning_rate_schedule='discexp')
 
 # Create trainer
 trainer = paddle.trainer.SGD(cost=cost,
@@ -475,7 +474,7 @@ Test with Pass 0, {'classification_error_evaluator': 0.885200023651123}
 
 ## 应用模型
 
-可以使用训练好的模型对图片进行分类，下面程序展示了如何使用`paddle.infer`接口进行推断。
+可以使用训练好的模型对图片进行分类，下面程序展示了如何使用`paddle.infer`接口进行推断，可以打开注释，更改加载的模型。
 
 ```python
 from PIL import Image
@@ -484,13 +483,24 @@ import os
 def load_image(file):
     im = Image.open(file)
     im = im.resize((32, 32), Image.ANTIALIAS)
-    im = np.array(im).astype(np.float32).flatten()
+    im = np.array(im).astype(np.float32)
+    # PIL打开图片存储顺序为H(高度)，W(宽度)，C(通道)。
+    # PaddlePaddle要求数据顺序为CHW，所以需要转换顺序。
+    im = im.transpose((2, 0, 1)) # CHW
+    # CIFAR训练图片通道顺序为B(蓝),G(绿),R(红),
+    # 而PIL打开图片默认通道顺序为RGB,因为需要交换通道。
+    im = im[(2, 1, 0),:,:] # BGR
+    im = im.flatten()
     im = im / 255.0
     return im
+
 test_data = []
 cur_dir = os.path.dirname(os.path.realpath(__file__))
 test_data.append((load_image(cur_dir + '/image/dog.png'),)
 
+# with gzip.open('params_pass_50.tar.gz', 'r') as f:
+#    parameters = paddle.parameters.Parameters.from_tar(f)
+
 probs = paddle.infer(
     output_layer=out, parameters=parameters, input=test_data)
 lab = np.argsort(-probs) # probs and lab are the results of one batch data

03.image_classification/index.en.html

@@ -406,8 +406,7 @@ momentum_optimizer = paddle.optimizer.Momentum(
     learning_rate=0.1 / 128.0,
     learning_rate_decay_a=0.1,
     learning_rate_decay_b=50000 * 100,
-    learning_rate_schedule='discexp',
-    batch_size=128)
+    learning_rate_schedule='discexp')
 
 # Create trainer
 trainer = paddle.trainer.SGD(cost=cost,
@@ -525,7 +524,7 @@ Figure 12. The error rate of VGG model on CIFAR10
 
 ## Application
 
-After training is done, users can use the trained model to classify images. The following code shows how to infer through `paddle.infer` interface.
+After training is done, users can use the trained model to classify images. The following code shows how to infer through `paddle.infer` interface. You can remove the comments to change the model name.
 
 ```python
 from PIL import Image
@@ -534,13 +533,26 @@ import os
 def load_image(file):
     im = Image.open(file)
     im = im.resize((32, 32), Image.ANTIALIAS)
-    im = np.array(im).astype(np.float32).flatten()
+    im = np.array(im).astype(np.float32)
+    # The storage order of the loaded image is W(widht),
+    # H(height), C(channel). PaddlePaddle requires
+    # the CHW order, so transpose them.
+    im = im.transpose((2, 0, 1)) # CHW
+    # In the training phase, the channel order of CIFAR
+    # image is B(Blue), G(green), R(Red). But PIL open
+    # image in RGB mode. It must swap the channel order.
+    im = im[(2, 1, 0),:,:] # BGR
+    im = im.flatten()
     im = im / 255.0
     return im
 test_data = []
 cur_dir = os.path.dirname(os.path.realpath(__file__))
 test_data.append((load_image(cur_dir + '/image/dog.png'),)
 
+# users can remove the comments and change the model name
+# with gzip.open('params_pass_50.tar.gz', 'r') as f:
+#    parameters = paddle.parameters.Parameters.from_tar(f)
+
 probs = paddle.infer(
     output_layer=out, parameters=parameters, input=test_data)
 lab = np.argsort(-probs) # probs and lab are the results of one batch data

03.image_classification/index.html

@@ -398,8 +398,7 @@ momentum_optimizer = paddle.optimizer.Momentum(
     learning_rate=0.1 / 128.0,
     learning_rate_decay_a=0.1,
     learning_rate_decay_b=50000 * 100,
-    learning_rate_schedule='discexp',
-    batch_size=128)
+    learning_rate_schedule='discexp')
 
 # Create trainer
 trainer = paddle.trainer.SGD(cost=cost,
@@ -517,7 +516,7 @@ Test with Pass 0, {'classification_error_evaluator': 0.885200023651123}
 
 ## 应用模型
 
-可以使用训练好的模型对图片进行分类，下面程序展示了如何使用`paddle.infer`接口进行推断。
+可以使用训练好的模型对图片进行分类，下面程序展示了如何使用`paddle.infer`接口进行推断，可以打开注释，更改加载的模型。
 
 ```python
 from PIL import Image
@@ -526,13 +525,24 @@ import os
 def load_image(file):
     im = Image.open(file)
     im = im.resize((32, 32), Image.ANTIALIAS)
-    im = np.array(im).astype(np.float32).flatten()
+    im = np.array(im).astype(np.float32)
+    # PIL打开图片存储顺序为H(高度)，W(宽度)，C(通道)。
+    # PaddlePaddle要求数据顺序为CHW，所以需要转换顺序。
+    im = im.transpose((2, 0, 1)) # CHW
+    # CIFAR训练图片通道顺序为B(蓝),G(绿),R(红),
+    # 而PIL打开图片默认通道顺序为RGB,因为需要交换通道。
+    im = im[(2, 1, 0),:,:] # BGR
+    im = im.flatten()
     im = im / 255.0
     return im
+
 test_data = []
 cur_dir = os.path.dirname(os.path.realpath(__file__))
 test_data.append((load_image(cur_dir + '/image/dog.png'),)
 
+# with gzip.open('params_pass_50.tar.gz', 'r') as f:
+#    parameters = paddle.parameters.Parameters.from_tar(f)
+
 probs = paddle.infer(
     output_layer=out, parameters=parameters, input=test_data)
 lab = np.argsort(-probs) # probs and lab are the results of one batch data

03.image_classification/train.py

@@ -54,8 +54,7 @@ def main():
         learning_rate=0.1 / 128.0,
         learning_rate_decay_a=0.1,
         learning_rate_decay_b=50000 * 100,
-        learning_rate_schedule='discexp',
-        batch_size=128)
+        learning_rate_schedule='discexp')
 
     # End batch and end pass event handler
     def event_handler(event):
@@ -86,7 +85,7 @@ def main():
             paddle.reader.shuffle(
                 paddle.dataset.cifar.train10(), buf_size=50000),
             batch_size=128),
-        num_passes=1,
+        num_passes=200,
         event_handler=event_handler,
         feeding={'image': 0,
                  'label': 1})
@@ -99,7 +98,16 @@ def main():
     def load_image(file):
         im = Image.open(file)
         im = im.resize((32, 32), Image.ANTIALIAS)
-        im = np.array(im).astype(np.float32).flatten()
+        im = np.array(im).astype(np.float32)
+        # The storage order of the loaded image is W(widht),
+        # H(height), C(channel). PaddlePaddle requires
+        # the CHW order, so transpose them.
+        im = im.transpose((2, 0, 1))  # CHW
+        # In the training phase, the channel order of CIFAR
+        # image is B(Blue), G(green), R(Red). But PIL open
+        # image in RGB mode. It must swap the channel order.
+        im = im[(2, 1, 0), :, :]  # BGR
+        im = im.flatten()
         im = im / 255.0
         return im
 
@@ -107,6 +115,10 @@ def main():
     cur_dir = os.path.dirname(os.path.realpath(__file__))
     test_data.append((load_image(cur_dir + '/image/dog.png'), ))
 
+    # users can remove the comments and change the model name
+    # with gzip.open('params_pass_50.tar.gz', 'r') as f:
+    #    parameters = paddle.parameters.Parameters.from_tar(f)
+
     probs = paddle.infer(
         output_layer=out, parameters=parameters, input=test_data)
     lab = np.argsort(-probs)  # probs and lab are the results of one batch data