update docs

examples/basics/mnist-tflayers.py

@@ -44,20 +44,17 @@ class Model(ModelDesc):
 
         image = image * 2 - 1   # center the pixels values at zero
 
-        # The context manager `argscope` sets the default option for all the layers under
-        # this context. Here we use 32 channel convolution with shape 3x3
-        with argscope(Conv2D, kernel_shape=3, nl=tf.nn.relu, out_channel=32):
-            l = tf.layers.conv2d(image, 32, 3, padding='same', activation=tf.nn.relu, name='conv0')
-            l = tf.layers.max_pooling2d(l, 2, 2, padding='valid')
-            l = tf.layers.conv2d(l, 32, 3, padding='same', activation=tf.nn.relu, name='conv1')
-            l = tf.layers.conv2d(l, 32, 3, padding='same', activation=tf.nn.relu, name='conv2')
-            l = tf.layers.max_pooling2d(l, 2, 2, padding='valid')
-            l = tf.layers.conv2d(l, 32, 3, padding='same', activation=tf.nn.relu, name='conv3')
-            l = tf.layers.flatten(l)
-            l = tf.layers.dense(l, 512, activation=tf.nn.relu, name='fc0')
-            l = tf.layers.dropout(l, rate=0.5,
-                                  training=get_current_tower_context().is_training)
-            logits = tf.layers.dense(l, 10, activation=tf.identity, name='fc1')
+        l = tf.layers.conv2d(image, 32, 3, padding='same', activation=tf.nn.relu, name='conv0')
+        l = tf.layers.max_pooling2d(l, 2, 2, padding='valid')
+        l = tf.layers.conv2d(l, 32, 3, padding='same', activation=tf.nn.relu, name='conv1')
+        l = tf.layers.conv2d(l, 32, 3, padding='same', activation=tf.nn.relu, name='conv2')
+        l = tf.layers.max_pooling2d(l, 2, 2, padding='valid')
+        l = tf.layers.conv2d(l, 32, 3, padding='same', activation=tf.nn.relu, name='conv3')
+        l = tf.layers.flatten(l)
+        l = tf.layers.dense(l, 512, activation=tf.nn.relu, name='fc0')
+        l = tf.layers.dropout(l, rate=0.5,
+                              training=get_current_tower_context().is_training)
+        logits = tf.layers.dense(l, 10, activation=tf.identity, name='fc1')
 
         tf.nn.softmax(logits, name='prob')   # a Bx10 with probabilities
 

tensorpack/callbacks/graph.py

@@ -100,6 +100,8 @@ class ProcessTensors(Callback):
     """
     Fetch extra tensors **along with** each training step,
     and call some function over the values.
+    It uses `_{before,after}_run` method to inject `tf.train.SessionRunHooks`
+    to the session.
     You can use it to print tensors, save tensors to file, etc.
 
     Examples:

tensorpack/callbacks/trigger.py

@@ -11,6 +11,9 @@ __all__ = ['PeriodicTrigger', 'PeriodicRunHooks', 'EnableCallbackIf']
 class PeriodicTrigger(ProxyCallback):
     """
     Schedule to trigger a callback every k global steps or every k epochs by its ``trigger()`` method.
+
+    Note that it does not touch other methods (``before/after_run``,
+    ``trigger_step``, etc).
     """
 
     _chief_only = False
@@ -99,7 +102,7 @@ class EnableCallbackIf(ProxyCallback):
         """
         Args:
             callback (Callback):
-            pred (self -> bool): a callable predicate
+            pred (self -> bool): a callable predicate. Has to be a pure function.
         """
         self._pred = pred
         super(EnableCallbackIf, self).__init__(callback)
@@ -134,3 +137,6 @@ class EnableCallbackIf(ProxyCallback):
     def _trigger_step(self):
         if self._pred(self):
             super(EnableCallbackIf, self)._trigger_step()
+
+    def __str__(self):
+        return "EnableCallbackIf-" + str(self.cb)