specify the names of all layers to adapt in configs and trainer

bob/learn/pytorch/config/mccnn/wmca_mccnn.py

@@ -87,7 +87,7 @@ learning_rate=0.0001
 seed = 3
 output_dir = 'training_mccn'
 use_gpu = False
-adapted_layers = 'conv1-block1'
+adapted_layers = 'conv1-ffc'
 adapt_reference_channel = False
 verbose = 2
 

bob/learn/pytorch/scripts/train_mccnn.py

@@ -19,7 +19,7 @@ Options:
       --batch-size=<int>                Batch size [default: 64]
       --epochs=<int>                    Number of training epochs [default: 20]
       --learning-rate=<float>           Learning rate [default: 0.01]
-      --adapted-layers=<string>         Layers to adapt in the training [default: conv1-block1-group1]
+      --adapted-layers=<string>         Layers to adapt in the training [default: conv1-block1-group1-ffc]
       --adapt-reference-channel         Flag deciding whether to adapt default channel as well [default: False]
   -S, --seed=<int>                      The random seed [default: 3] 
   -o, --output-dir=<path>               Dir to save stuff [default: training]
@@ -82,7 +82,7 @@ def main(user_input=None):
   output_dir = get_parameter(args, configuration, 'output_dir', 'training')
   use_gpu = get_parameter(args, configuration, 'use_gpu', False)
   verbosity_level = get_parameter(args, configuration, 'verbose', 0)
-  adapted_layers = get_parameter(args, configuration, 'adapted_layers', 'conv1-block1-group1')
+  adapted_layers = get_parameter(args, configuration, 'adapted_layers', 'conv1-block1-group1-ffc')
   adapt_reference_channel = get_parameter(args, configuration, 'adapt_reference_channel', False)
   
   bob.core.log.set_verbosity_level(logger, verbosity_level)

bob/learn/pytorch/trainers/MCCNNTrainer.py

@@ -72,7 +72,7 @@ class MCCNNTrainer(object):
   
   """
  
-  def __init__(self, network, batch_size=64, use_gpu=False, adapted_layers='conv1-block1-group1',adapt_reference_channel=False, verbosity_level=2):
+  def __init__(self, network, batch_size=64, use_gpu=False, adapted_layers='conv1-block1-group1-ffc',adapt_reference_channel=False, verbosity_level=2):
     """ Init function . The layers to be adapted in the network is selected and the gradients are set to `True` 
     for the  layers which needs to be adapted. 
 
@@ -86,7 +86,7 @@ class MCCNNTrainer(object):
       If you would like to use the gpu
     adapted_layers: str
       The blocks in the CNN to adapt; only the ones listed are adapted in the training. The layers are separated by '-' in the
-      string, for example 'conv1-block1-group1'. The fully connected layer in the output part are adapted always.
+      string, for example 'conv1-block1-group1-ffc'. The fully connected layer in the output part are adapted always.
     adapt_reference_channel: bool
       If this value is `True` then 'ch_0' (which is the reference channel- usually, grayscale image) is also adapted. Otherwise the reference channel
       is not adapted, so that it can be used for Face recognition as well, default: `False`.
@@ -106,10 +106,14 @@ class MCCNNTrainer(object):
 
     layers_present = self.network.lcnn_layers
 
+    layers_present.append('ffc')
+
     # select the layers in the network to adapt
 
     adapted_layers_list=adapted_layers.split('-')
 
+    assert('ffc' in adapted_layers_list)
+
     assert(set(adapted_layers_list)<=set(layers_present)) # to ensure layer names are valid
 
     if adapt_reference_channel:# whether to adapt the color channel
@@ -124,8 +128,9 @@ class MCCNNTrainer(object):
     for i in range(start_index,self.network.num_channels):
 
       for layer in adapted_layers_list:
+          if layer!='ffc':
 
-        layers_to_adapt.append("layer_dict.ch_{}_".format(i)+layer)
+           layers_to_adapt.append("layer_dict.ch_{}_".format(i)+layer)
 
     layers_to_adapt = list(np.unique(layers_to_adapt))