Updates

• config file from train.py accepts tensors as input
• TFRecord, add keyword argument to flag the shuffling mechanism
• Added InceptionResnetV1 in our basket of networks
• Added an option to sum the regularization losses in the CrossEntropy loss
• Implemented validation with embeddings (officially appending to the tensorboard)
• Created method Trainer.load_variables_from_external_model that load variables from another model in the current graph
• Implemented MeanSoftMaxLossCenterLoss (http://ydwen.github.io/papers/WenECCV16.pdf)
• Patched InceptionV2 (Added the flag trainable for each variable in the model)
• Implemented TFRecordImage (read, data augmentation, scaling and casting to float32)
• Added data type as input in the TFRecord datashufflers

bob/learn/tensorflow/trainers/Trainer.py

@@ -60,7 +60,7 @@ class Trainer(object):
                  ###### training options ##########
                  iterations=5000,
                  snapshot=1000,
-                 validation_snapshot=2000,
+                 validation_snapshot=2000,#2000,
                  keep_checkpoint_every_n_hours=2,
 
                  ## Analizer
@@ -421,6 +421,7 @@ class Trainer(object):
             step: Iteration number
 
         """
+        
         if self.validation_data_shuffler.prefetch:
             embedding, labels = self.session.run([self.validation_predictor, self.validation_label_ph])
         else: