update examples

Author: Laicheng0830

examples/basic_tutorials/tutorial_SequentialLayer.py

@@ -1,5 +1,7 @@
 #! /usr/bin/python
 # -*- coding: utf-8 -*-
+import os
+os.environ['TL_BACKEND'] = 'tensorflow'
 
 from tensorlayer.layers import SequentialLayer
 from tensorlayer.layers import Dense

examples/basic_tutorials/tutorial_automatic_inference_input _shape.py

@@ -1,5 +1,7 @@
 #! /usr/bin/python
 # -*- coding: utf-8 -*-
+import os
+os.environ['TL_BACKEND'] = 'tensorflow'
 
 import numpy as np
 import time

examples/basic_tutorials/tutorial_cifar10_cnn_tensorflow_backend.py

@@ -1,6 +1,8 @@
 #! /usr/bin/python
 # -*- coding: utf-8 -*-
 # The tensorlayer and tensorflow operators can be mixed
+import os
+os.environ['TL_BACKEND'] = 'tensorflow'
 
 import time
 import numpy as np

examples/basic_tutorials/tutorial_mnist_mlp_mindspore_backend.py

@@ -1,5 +1,7 @@
 #! /usr/bin/python
 # -*- coding: utf-8 -*-
+import os
+os.environ['TL_BACKEND'] = 'mindspore'
 
 import mindspore.ops.operations as P
 from mindspore.ops import composite as C

examples/basic_tutorials/tutorial_mnist_mlp_tensorflow_backend.py

@@ -1,6 +1,8 @@
 #! /usr/bin/python
 # -*- coding: utf-8 -*-
 # The tensorlayer and tensorflow operators can be mixed
+import os
+os.environ['TL_BACKEND'] = 'tensorflow'
 
 import numpy as np
 import time

examples/basic_tutorials/tutorial_mnist_simple.py

@@ -3,30 +3,27 @@
 
 # The same set of code can switch the backend with one line
 import os
-os.environ['TL_BACKEND'] = 'tensorflow'
+# os.environ['TL_BACKEND'] = 'tensorflow'
 # os.environ['TL_BACKEND'] = 'mindspore'
-# os.environ['TL_BACKEND'] = 'paddle'
+os.environ['TL_BACKEND'] = 'paddle'
 
 import tensorlayer as tl
 from tensorlayer.layers import Module
 from tensorlayer.layers import Dense, Dropout, Flatten
 from tensorlayer.dataflow import Dataset
-from tensorlayer.vision.transforms import Normalize, Compose
 
-X_train, y_train, X_val, y_val, X_test, y_test = tl.files.load_mnist_dataset(shape=(-1, 28, 28, 1))
 
-transform = Compose([Normalize(mean=[127.5/255.], std=[127.5/255.], data_format='HWC')])
+X_train, y_train, X_val, y_val, X_test, y_test = tl.files.load_mnist_dataset(shape=(-1, 784))
+
 
 class mnistdataset(Dataset):
 
-    def __init__(self, data = X_train, label = y_train ,transform = transform):
+    def __init__(self, data = X_train, label = y_train):
         self.data = data
         self.label = label
-        self.transform = transform
 
     def __getitem__(self, index):
         data = self.data[index].astype('float32')
-        data = self.transform(data)
         label = self.label[index].astype('int64')
 
         return data, label
@@ -46,11 +43,9 @@ def __init__(self):
         self.dense2 = Dense(n_units=800, act=tl.ReLU, in_channels=800)
         self.dropout3 = Dropout(keep=0.8)
         self.dense3 = Dense(n_units=10, act=tl.ReLU, in_channels=800)
-        self.flatten = Flatten()
 
     def forward(self, x, foo=None):
-        z = self.flatten(x)
-        z = self.dropout1(z)
+        z = self.dropout1(x)
         z = self.dense1(z)
         z = self.dropout2(z)
         z = self.dense2(z)
@@ -68,9 +63,9 @@ def forward(self, x, foo=None):
 
 
 train_weights = MLP.trainable_weights
-optimizer = tl.optimizers.Momentum(0.001, 0.9)
+optimizer = tl.optimizers.Momentum(0.05, 0.9)
 metric = tl.metric.Accuracy()
-train_dataset = mnistdataset(data = X_train, label = y_train ,transform = transform)
+train_dataset = mnistdataset(data = X_train, label = y_train)
 train_dataset = tl.dataflow.FromGenerator(train_dataset, output_types=[tl.float32, tl.int64], column_names=['data', 'label'])
 train_loader = tl.dataflow.Dataloader(train_dataset, batch_size=batch_size, shuffle=True)
 

examples/basic_tutorials/tutorial_nested_usage_of_Layer.py

@@ -1,5 +1,7 @@
 #! /usr/bin/python
 # -*- coding: utf-8 -*-
+import os
+os.environ['TL_BACKEND'] = 'tensorflow'
 
 import time
 import numpy as np

tensorlayer/cost/mindspore_cost.py

@@ -26,33 +26,7 @@
 ]
 
 
-def softmax_cross_entropy_with_logits(output, target):
-    """Softmax cross-entropy operation, returns the TensorFlow expression of cross-entropy for two distributions,
-    it implements softmax internally. See ``tf.ops.sparse_softmax_cross_entropy_with_logits``.
-
-    Parameters
-    ----------
-    output : Tensor
-        A batch of distribution with shape: [batch_size, num of classes].
-    target : Tensor
-        A batch of index with shape: [batch_size, ].
-    name : string
-        Name of this loss.
-
-    Examples
-    --------
-    >>> import tensorlayer as tl
-    >>> ce = tl.cost.softmax_cross_entropy_with_logits(y_logits, y_target_logits, 'my_loss')
-
-    References
-    -----------
-    - About cross-entropy: `<https://en.wikipedia.org/wiki/Cross_entropy>`__.
-    - The code is borrowed from: `<https://en.wikipedia.org/wiki/Cross_entropy>`__.
-
-    """
-
-    outputs = nn.SoftmaxCrossEntropyWithLogits(sparse=True, reduction='mean')(output, target)
-    return outputs
+softmax_cross_entropy_with_logits = nn.SoftmaxCrossEntropyWithLogits(sparse=True, reduction='mean')
 
 
 def sigmoid_cross_entropy(output, target, name=None):