Many changes

• Removing legacy code
• Port to the new ci
• Estimate the size of the tfrecords file that will be created
• bug-fixes
• Add a new architecture for patch-based CNN
• Refactored the scripts to use the new click tools
• Changed the compute_statistics script completely
• Removed the unused predict_generic script
• Removed the unused train script and renamed train_generic to train

Fixes #50 (closed) #52 (closed) #48 (closed)

bob/learn/tensorflow/network/PatchCNN.py

@@ -44,7 +44,8 @@ from __future__ import print_function
 import tensorflow as tf
 
 
-def base_architecture(input_layer, mode, data_format, **kwargs):
+def base_architecture(input_layer, mode, data_format,
+                      skip_first_two_pool=False, **kwargs):
     # Keep track of all the endpoints
     endpoints = {}
     bn_axis = 1 if data_format.lower() == 'channels_first' else 3
@@ -69,8 +70,12 @@ def base_architecture(input_layer, mode, data_format, **kwargs):
     endpoints['BN-1-activation'] = bn1_act
 
     # Pooling Layer #1
-    pool1 = tf.layers.max_pooling2d(
-        inputs=bn1_act, pool_size=[2, 2], strides=2, data_format=data_format)
+    if skip_first_two_pool:
+        pool1 = bn1_act
+    else:
+        pool1 = tf.layers.max_pooling2d(
+            inputs=bn1_act, pool_size=[2, 2], strides=2,
+            data_format=data_format)
     endpoints['MaxPooling-1'] = pool1
 
     # ======================
@@ -92,8 +97,12 @@ def base_architecture(input_layer, mode, data_format, **kwargs):
     endpoints['BN-2-activation'] = bn2_act
 
     # Pooling Layer #2
-    pool2 = tf.layers.max_pooling2d(
-        inputs=bn2_act, pool_size=[2, 2], strides=2, data_format=data_format)
+    if skip_first_two_pool:
+        pool2 = bn2_act
+    else:
+        pool2 = tf.layers.max_pooling2d(
+            inputs=bn2_act, pool_size=[2, 2], strides=2,
+            data_format=data_format)
     endpoints['MaxPooling-2'] = pool2
 
     # ======================
@@ -177,7 +186,7 @@ def base_architecture(input_layer, mode, data_format, **kwargs):
 
     # Batch Normalization #6
     bn6 = tf.layers.batch_normalization(
-        fc_1, axis=bn_axis, training=training, fused=True)
+        fc_1, axis=1, training=training, fused=True)
     endpoints['BN-6'] = bn6
     bn6_act = tf.nn.relu(bn6)
     endpoints['BN-6-activation'] = bn6_act
@@ -193,7 +202,7 @@ def base_architecture(input_layer, mode, data_format, **kwargs):
 
     # Batch Normalization #7
     bn7 = tf.layers.batch_normalization(
-        fc_2, axis=bn_axis, training=training, fused=True)
+        fc_2, axis=1, training=training, fused=True)
     endpoints['BN-7'] = bn7
     bn7_act = tf.nn.relu(bn7)
     endpoints['BN-7-activation'] = bn7_act
@@ -203,6 +212,7 @@ def base_architecture(input_layer, mode, data_format, **kwargs):
 
 def architecture(input_layer,
                  mode=tf.estimator.ModeKeys.TRAIN,
+                 skip_first_two_pool=False,
                  n_classes=2,
                  data_format='channels_last',
                  reuse=False,
@@ -210,7 +220,8 @@ def architecture(input_layer,
 
     with tf.variable_scope('PatchCNN', reuse=reuse):
 
-        bn7_act, endpoints = base_architecture(input_layer, mode, data_format)
+        bn7_act, endpoints = base_architecture(
+            input_layer, mode, data_format, skip_first_two_pool)
         # Logits layer
         logits = tf.layers.dense(inputs=bn7_act, units=n_classes)
         endpoints['FC-3'] = logits
@@ -229,8 +240,9 @@ def model_fn(features, labels, mode, params=None, config=None):
     momentum = params.get('momentum', 0.99)
 
     arch_kwargs = {
-        'n_classes': params.get('n_classes', None),
-        'data_format': params.get('data_format', None),
+        'skip_first_two_pool': params.get('skip_first_two_pool'),
+        'n_classes': params.get('n_classes'),
+        'data_format': params.get('data_format'),
     }
     arch_kwargs = {k: v for k, v in arch_kwargs.items() if v is not None}