add shape summary

example_mnist.py

@@ -35,15 +35,15 @@ def get_model(inputs):
     keep_prob = tf.get_default_graph().get_tensor_by_name(DROPOUT_PROB_VAR_NAME)
 
     image, label = inputs
+    image = tf.expand_dims(image, 3)    # add a single channel
 
-    image = tf.expand_dims(image, 3)
-    conv0 = Conv2D('conv0', image, out_channel=32, kernel_shape=5,
-                  padding='valid')
+    conv0 = Conv2D('conv0', image, out_channel=32, kernel_shape=5)
     pool0 = MaxPooling('pool0', conv0, 2)
     conv1 = Conv2D('conv1', pool0, out_channel=40, kernel_shape=3)
     pool1 = MaxPooling('pool1', conv1, 2)
+    conv2 = Conv2D('conv2', pool1, out_channel=32, kernel_shape=3)
 
-    fc0 = FullyConnected('fc0', pool1, 1024)
+    fc0 = FullyConnected('fc0', conv2, 1024)
     fc0 = tf.nn.dropout(fc0, keep_prob)
 
     # fc will have activation summary by default. disable this for the output layer

models/_common.py

@@ -5,6 +5,7 @@
 
 import tensorflow as tf
 from utils.summary import *
+from utils import logger
 
 def layer_register(summary_activation=False):
     """
@@ -17,15 +18,28 @@ def layer_register(summary_activation=False):
             args = args[1:]
             do_summary = kwargs.pop(
                 'summary_activation', summary_activation)
+            inputs = args[0]
+            if isinstance(inputs, list):
+                shape_str = ",".join(
+                    map(str(x.get_shape().as_list()), inputs))
+            else:
+                shape_str = str(inputs.get_shape().as_list())
+            logger.info("{} input: {}".format(name, shape_str))
 
             with tf.variable_scope(name) as scope:
-                ret = func(*args, **kwargs)
+                outputs = func(*args, **kwargs)
+                if isinstance(outputs, list):
+                    shape_str = ",".join(
+                        map(str(x.get_shape().as_list()), outputs))
                     if do_summary:
-                    ndim = ret.get_shape().ndims
-                    assert ndim >= 2, \
-                        "Summary a scalar with histogram? Maybe use scalar instead. FIXME!"
-                    add_activation_summary(ret, scope.name)
-                return ret
+                        for x in outputs:
+                            add_activation_summary(x, scope.name)
+                else:
+                    shape_str = str(outputs.get_shape().as_list())
+                    if do_summary:
+                        add_activation_summary(outputs, scope.name)
+                logger.info("{} output: {}".format(name, shape_str))
+                return outputs
         return inner
     return wrapper
 
@@ -35,7 +49,7 @@ def shape2d(a):
     """
     if type(a) == int:
         return [a, a]
-    if type(a) in [list, tuple]:
+    if isinstance(a, (list, tuple)):
         assert len(a) == 2
         return list(a)
     raise RuntimeError("Illegal shape: {}".format(a))

train.py

@@ -42,13 +42,13 @@ def start_train(config):
 
     max_epoch = int(config['max_epoch'])
 
-
     # build graph
     G = tf.get_default_graph()
     for v in input_vars:
         G.add_to_collection(INPUT_VARS_KEY, v)
     for v in output_vars:
         G.add_to_collection(OUTPUT_VARS_KEY, v)
+    summary_model()
 
     global_step_var = G.get_tensor_by_name(GLOBAL_STEP_VAR_NAME)
 

utils/__init__.py

@@ -9,6 +9,7 @@ import time
 import sys
 from contextlib import contextmanager
 import logger
+import tensorflow as tf
 
 def global_import(name):
     p = __import__(name, globals(), locals())
@@ -29,3 +30,18 @@ def timed_operation(msg, log_start=False):
     yield
     logger.info('finished {}, time={:.2f}sec.'.format(
         msg, time.time() - start))
+
+
+def summary_model():
+    train_vars = tf.get_collection(tf.GraphKeys.TRAINABLE_VARIABLES)
+    msg = [""]
+    total = 0
+    for v in train_vars:
+        shape = v.get_shape()
+        ele = shape.num_elements()
+        total += ele
+        msg.append("{}: shape={}, dim={}".format(
+            v.name, shape.as_list(), ele))
+    msg.append("Total dim={}".format(total))
+    logger.info("Model Params: {}".format('\n'.join(msg)))
+

utils/summary.py

@@ -22,6 +22,9 @@ def add_activation_summary(x, name=None):
     Summary for an activation tensor x.
     If name is None, use x.name
     """
+    ndim = x.get_shape().ndims
+    assert ndim >= 2, \
+        "Summary a scalar with histogram? Maybe use scalar instead. FIXME!"
     if name is None:
         name = x.name
     tf.histogram_summary(name + '/activations', x)