Siamese net with tf-slim

bob/learn/tensorflow/network/Chopra.py

@@ -67,7 +67,6 @@ class Chopra(object):
                  device="/cpu:0",
                  batch_norm=False):
 
-
             self.conv1_kernel_size = conv1_kernel_size
             self.conv1_output = conv1_output
             self.pooling1_size = pooling1_size
@@ -84,7 +83,6 @@ class Chopra(object):
 
     def __call__(self, inputs):
         slim = tf.contrib.slim
-        initializer = tf.contrib.layers.xavier_initializer(uniform=False, dtype=tf.float32, seed=self.seed)
 
         with tf.device(self.device):
 
@@ -106,6 +104,6 @@ class Chopra(object):
 
             graph = slim.fully_connected(graph, self.fc1_output,
                                          weights_initializer=initializer,
+                                         activation_fn=None,
                                          scope='fc1')
-
         return graph

bob/learn/tensorflow/test/test_cnn.py

@@ -104,7 +104,6 @@ def test_cnn_trainer():
     graph = architecture(inputs['data'])
     embedding = Embedding(inputs['data'], graph)
 
-
     # One graph trainer
     trainer = Trainer(inputs=inputs,
                       graph=graph,
@@ -113,15 +112,15 @@ def test_cnn_trainer():
                       analizer=None,
                       prefetch=False,
                       learning_rate=constant(0.01, name="regular_lr"),
+                      optimizer=tf.train.GradientDescentOptimizer(0.01),
                       temp_dir=directory
                       )
     trainer.train(train_data_shuffler)
     accuracy = validate_network(embedding, validation_data, validation_labels)
-    #import ipdb; ipdb.set_trace()
 
     # At least 80% of accuracy
     assert accuracy > 80.
-    #shutil.rmtree(directory)
+    shutil.rmtree(directory)
     del trainer
     del graph
 
@@ -143,12 +142,21 @@ def test_siamesecnn_trainer():
 
     # Preparing the architecture
     architecture = Chopra(seed=seed, fc1_output=10)
+    inputs = {}
+    inputs['left'] = tf.placeholder(tf.float32, shape=[None, 28, 28, 1], name="input_left")
+    inputs['right'] = tf.placeholder(tf.float32, shape=[None, 28, 28, 1], name="input_right")
+    inputs['label'] = tf.placeholder(tf.int64, shape=[None], name="label")
+
+    graph = {}
+    graph['left'] = architecture(inputs['left'])
+    graph['right'] = architecture(inputs['right'])
 
     # Loss for the Siamese
     loss = ContrastiveLoss(contrastive_margin=4.)
 
     # One graph trainer
-    trainer = SiameseTrainer(architecture=architecture,
+    trainer = SiameseTrainer(inputs=inputs,
+                             graph=graph,
                              loss=loss,
                              iterations=iterations,
                              prefetch=False,
@@ -158,6 +166,7 @@ def test_siamesecnn_trainer():
                              temp_dir=directory
                              )
 
+    import ipdb; ipdb.set_trace();
     trainer.train(train_data_shuffler)
 
     eer = dummy_experiment(validation_data_shuffler, architecture)

bob/learn/tensorflow/trainers/SiameseTrainer.py

@@ -8,6 +8,7 @@ from tensorflow.core.framework import summary_pb2
 from ..analyzers import ExperimentAnalizer, SoftmaxAnalizer
 from ..network import SequenceNetwork
 from .Trainer import Trainer
+from .learning_rate import constant
 import os
 import logging
 logger = logging.getLogger("bob.learn")
@@ -60,7 +61,8 @@ class SiameseTrainer(Trainer):
     """
 
     def __init__(self,
-                 architecture,
+                 inputs,
+                 graph,
                  optimizer=tf.train.AdamOptimizer(),
                  use_gpu=False,
                  loss=None,
@@ -84,30 +86,56 @@ class SiameseTrainer(Trainer):
                  verbosity_level=2
                  ):
 
-        super(SiameseTrainer, self).__init__(
-            architecture=architecture,
-            optimizer=optimizer,
-            use_gpu=use_gpu,
-            loss=loss,
-            temp_dir=temp_dir,
+        import ipdb;
+        ipdb.set_trace();
 
-            # Learning rate
-            learning_rate=learning_rate,
+        self.inputs = inputs
+        self.graph = graph
+        self.loss = loss
 
-            ###### training options ##########
-            convergence_threshold=convergence_threshold,
-            iterations=iterations,
-            snapshot=snapshot,
-            validation_snapshot=validation_snapshot,
-            prefetch=prefetch,
+        if not isinstance(self.graph, dict) or not(('left' and 'right') in self.graph.keys()):
+            raise ValueError("Expected a dict with the elements `right` and `left` as input for the keywork `graph`")
 
-            ## Analizer
-            analizer=analizer,
+        self.predictor = self.loss(self.graph, inputs['label'])
 
-            model_from_file=model_from_file,
+        self.optimizer_class = optimizer
+        self.use_gpu = use_gpu
+        self.temp_dir = temp_dir
 
-            verbosity_level=verbosity_level
-        )
+        if learning_rate is None and model_from_file == "":
+            self.learning_rate = constant()
+        else:
+            self.learning_rate = learning_rate
+
+        self.iterations = iterations
+        self.snapshot = snapshot
+        self.validation_snapshot = validation_snapshot
+        self.convergence_threshold = convergence_threshold
+        self.prefetch = prefetch
+
+        # Training variables used in the fit
+        self.optimizer = None
+        self.training_graph = None
+        self.train_data_shuffler = None
+        self.summaries_train = None
+        self.train_summary_writter = None
+        self.thread_pool = None
+
+        # Validation data
+        self.validation_graph = None
+        self.validation_summary_writter = None
+
+        # Analizer
+        self.analizer = analizer
+
+        self.thread_pool = None
+        self.enqueue_op = None
+        self.global_step = None
+
+        self.model_from_file = model_from_file
+        self.session = None
+
+        bob.core.log.set_verbosity_level(logger, verbosity_level)
 
         self.between_class_graph_train = None
         self.within_class_graph_train = None
@@ -115,6 +143,7 @@ class SiameseTrainer(Trainer):
         self.between_class_graph_validation = None
         self.within_class_graph_validation = None
 
+
     def bootstrap_placeholders(self, train_data_shuffler, validation_data_shuffler):
         """
         Persist the placeholders

bob/learn/tensorflow/trainers/Trainer.py

@@ -94,12 +94,13 @@ class Trainer(object):
 
                  verbosity_level=2):
 
-        #if not isinstance(graph, SequenceNetwork):
-        #    raise ValueError("`architecture` should be instance of `SequenceNetwork`")
-
         self.inputs = inputs
         self.graph = graph
         self.loss = loss
+
+        if not isinstance(self.graph, tf.Tensor):
+            raise ValueError("Expected a tf.Tensor as input for the keywork `graph`")
+
         self.predictor = self.loss(self.graph, inputs['label'])
 
         self.optimizer_class = optimizer
@@ -325,10 +326,6 @@ class Trainer(object):
         self.train_data_shuffler = train_data_shuffler
 
         logger.info("Initializing !!")
-
-        if not isinstance(self.graph, tf.Tensor):
-            raise NotImplemented("Not tensor still not implemented")
-
         self.session = Session.instance(new=True).session
 
         # Loading a pretrained model