add simple mnist

docs/user/installation.rst

@@ -68,7 +68,7 @@ TensorFlow
 -----------
 
 `TensorFlow <https://www.tensorflow.org/versions/master/get_started/os_setup.html>`_ only officially
-supports GPU acceleration for Linux at present.
+supports GPU acceleration for Linux and Mac OX at present.
 If you want to use GPU with Mac OX, you need to compile TensorFlow from source.
 
 .. warning::

tutorial_mnist.py

@@ -1,11 +1,10 @@
 #! /usr/bin/python
 # -*- coding: utf8 -*-
 
-
+import numpy as np
 import tensorflow as tf
 import tensorlayer as tl
 from tensorlayer.layers import set_keep
-import numpy as np
 import time
 
 """Examples of Stacked Denoising Autoencoder, Dropout, Dropconnect and CNN.
@@ -36,11 +35,11 @@ def main_test_layers(model='relu'):
                                     tl.files.load_mnist_dataset(shape=(-1,784))
 
     X_train = np.asarray(X_train, dtype=np.float32)
-    y_train = np.asarray(y_train, dtype=np.int64)
+    y_train = np.asarray(y_train, dtype=np.int32)
     X_val = np.asarray(X_val, dtype=np.float32)
-    y_val = np.asarray(y_val, dtype=np.int64)
+    y_val = np.asarray(y_val, dtype=np.int32)
     X_test = np.asarray(X_test, dtype=np.float32)
-    y_test = np.asarray(y_test, dtype=np.int64)
+    y_test = np.asarray(y_test, dtype=np.int32)
 
     print('X_train.shape', X_train.shape)
     print('y_train.shape', y_train.shape)
@@ -54,7 +53,7 @@ def main_test_layers(model='relu'):
 
     # placeholder
     x = tf.placeholder(tf.float32, shape=[None, 784], name='x')
-    y_ = tf.placeholder(tf.int64, shape=[None, ], name='y_')
+    y_ = tf.placeholder(tf.int32, shape=[None, ], name='y_')
 
     if model == 'relu':
         network = tl.layers.InputLayer(x, name='input_layer')

tutorial_mnist_simple.py

@@ -0,0 +1,122 @@
+#! /usr/bin/python
+# -*- coding: utf8 -*-
+
+
+import tensorflow as tf
+import tensorlayer as tl
+from tensorlayer.layers import set_keep
+import numpy as np
+import time
+
+"""Examples of MLP.
+
+tensorflow (0.9.0)
+"""
+
+def main():
+    X_train, y_train, X_val, y_val, X_test, y_test = \
+                                    tl.files.load_mnist_dataset(shape=(-1,784))
+
+    X_train = np.asarray(X_train, dtype=np.float32)
+    y_train = np.asarray(y_train, dtype=np.int32)
+    X_val = np.asarray(X_val, dtype=np.float32)
+    y_val = np.asarray(y_val, dtype=np.int32)
+    X_test = np.asarray(X_test, dtype=np.float32)
+    y_test = np.asarray(y_test, dtype=np.int32)
+
+    print('X_train.shape', X_train.shape)
+    print('y_train.shape', y_train.shape)
+    print('X_val.shape', X_val.shape)
+    print('y_val.shape', y_val.shape)
+    print('X_test.shape', X_test.shape)
+    print('y_test.shape', y_test.shape)
+    print('X %s   y %s' % (X_test.dtype, y_test.dtype))
+
+
+    n_epoch = 200
+    batch_size = 500
+    learning_rate = 0.0001
+    print_freq = 10
+    is_val = False
+
+    sess = tf.InteractiveSession()
+
+    x = tf.placeholder(tf.float32, shape=[None, 784], name='x')
+    y_ = tf.placeholder(tf.int64, shape=[None, ], name='y_')
+
+    network = tl.layers.InputLayer(x, name='input_layer')
+    network = tl.layers.DropoutLayer(network, keep=0.8, name='drop1')
+    network = tl.layers.DenseLayer(network, n_units=800,
+                                    act = tf.nn.relu, name='relu1')
+    network = tl.layers.DropoutLayer(network, keep=0.5, name='drop2')
+    network = tl.layers.DenseLayer(network, n_units=800,
+                                    act = tf.nn.relu, name='relu2')
+    network = tl.layers.DropoutLayer(network, keep=0.5, name='drop3')
+    network = tl.layers.DenseLayer(network, n_units=10,
+                                    act = tl.activation.identity,
+                                    name='output_layer')
+
+    y = network.outputs
+    y_op = tf.argmax(tf.nn.softmax(y), 1)
+    cost = tf.reduce_mean(tf.nn.sparse_softmax_cross_entropy_with_logits(y, y_))
+
+    params = network.all_params
+
+
+    train_op = tf.train.AdamOptimizer(learning_rate, beta1=0.9, beta2=0.999,
+                                epsilon=1e-08, use_locking=False).minimize(cost)
+
+    sess.run(tf.initialize_all_variables())
+
+    network.print_params()
+    network.print_layers()
+
+    print('   learning_rate: %f' % learning_rate)
+    print('   batch_size: %d' % batch_size)
+
+    start_time_begin = time.time()
+    for epoch in range(n_epoch):
+        start_time = time.time()
+        loss_ep = 0; n_step = 0
+        for X_train_a, y_train_a in tl.iterate.minibatches(X_train, y_train,
+                                                    batch_size, shuffle=True):
+            feed_dict = {x: X_train_a, y_: y_train_a}
+            feed_dict.update( network.all_drop )    # enable dropout
+            loss, _ = sess.run([cost, train_op], feed_dict=feed_dict)
+            loss_ep += loss
+            n_step += 1
+        loss_ep = loss_ep/ n_step
+
+
+        if is_val:
+            if epoch + 1 == 1 or (epoch + 1) % print_freq == 0:
+                print("Epoch %d of %d took %fs" % (epoch + 1, n_epoch, time.time() - start_time))
+                dp_dict = tl.utils.dict_to_one( network.all_drop ) # disable dropout
+                feed_dict = {x: X_train, y_: y_train}
+                feed_dict.update(dp_dict)
+                print("   train loss: %f" % sess.run(cost, feed_dict=feed_dict))
+                dp_dict = tl.utils.dict_to_one( network.all_drop )
+                feed_dict = {x: X_val, y_: y_val}
+                feed_dict.update(dp_dict)
+                print("   val loss: %f" % sess.run(cost, feed_dict=feed_dict))
+                print("   val acc: %f" % np.mean(y_val ==
+                                        sess.run(y_op, feed_dict=feed_dict)))
+        else:
+            print("Epoch %d of %d took %fs, loss %f" % (epoch + 1, n_epoch, time.time() - start_time, loss_ep))
+    print("Total training time: %f" % (time.time() - start_time_begin))
+
+    print('Evaluation')
+    dp_dict = tl.utils.dict_to_one( network.all_drop )
+    feed_dict = {x: X_test, y_: y_test}
+    feed_dict.update(dp_dict)
+    print("   test loss: %f" % sess.run(cost, feed_dict=feed_dict))
+    print("   test acc: %f" % np.mean(y_test == sess.run(y_op,
+                                                        feed_dict=feed_dict)))
+
+    tl.files.save_npz(network.all_params , name='model.npz')
+    sess.close()
+
+
+if __name__ == '__main__':
+    sess = tf.InteractiveSession()
+    main()