small mods

doc/running_fasnet.md

@@ -12,8 +12,8 @@ FASNet accepts RGB images only, hence the preprocesing is done first. This can b
 
 .. code-block:: sh
 	./bin/spoof.py \
-	<PATH_TO_CONFIG>/wmca_grandtest_dbconfig.py \
-	<PATH_TO_CONFIG>/wmca_config_pytorch_extractor.py \
+	wmca-color \
+	fasnet \
 	--execute-only preprocessing \
 	--sub-directory <FASNET_PIPELINE_FOLDER> 
 	--grid idiap 
@@ -34,7 +34,7 @@ Once the config file is defined, training the network can be done with the follo
 .. code-block:: sh
 
     ./bin/train_fasnet.py \                   # script used for FASNET training
-    <PATH_TO_TRAINER_CONFIG>/wmca_FASNet.py \  # configuration file defining the FASNET network, database, and training parameters
+    <PATH_TO_TRAINER_CONFIG>/wmca_faset.py \  # configuration file defining the FASNET network, database, and training parameters
     -vv                                      # set verbosity level
 
 People in Idiap can benefit from GPU cluster, running the training as follows:
@@ -46,7 +46,7 @@ People in Idiap can benefit from GPU cluster, running the training as follows:
     --log-dir <FOLDER_TO_SAVE_THE_RESULTS>/logs/ \ # substitute the path to save the logs to (Idiap only)
     --environment="PYTHONUNBUFFERED=1" -- \        #
     ./bin/train_fasnet.py \                        # script used for FASNET training
-    <PATH_TO_TRAINER_CONFIG>/wmca_FASNet.py \      # configuration file defining the FASNET network, database, and training parameters
+    <PATH_TO_TRAINER_CONFIG>/wmca_fasnet.py \      # configuration file defining the FASNET network, database, and training parameters
     --use-gpu \                                    # enable the GPU mode
     -vv                                            # set verbosity level
 
@@ -57,7 +57,7 @@ For a more detailed documentation of functionality available in the training scr
 
     ./bin/train_fasnet.py --help   # note: remove ./bin/ if buildout is not used
 
-Please inspect the corresponding configuration file, ``wmca_FASNet.py`` for example, for more details on how to define the database, network architecture and training parameters.
+Please inspect the corresponding configuration file, ``wmca_faset.py`` for example, for more details on how to define the database, network architecture and training parameters.
 
 The protocols, and channels used in the experiments can be easily configured in the configuration file.
 
@@ -72,10 +72,10 @@ For **grandtest** protocol this can be done as follows.
 .. code-block:: sh
 
 	./bin/spoof.py \
-	<PATH_TO_DATABASE_CONFIG>/wmca_grandtest_dbconfig.py \
-	<PATH_TO_EXTRACTORS>/wmca_config_pytorch_extractor_v1_FASNet.py \
+	wmca-color \
+	fasnet \
 	--protocol grandtest \
-	--sub-directory <FOLDER_TO_SAVE_MCCNN_RESULTS>  -vv  
+	--sub-directory <FOLDER_TO_SAVE_FASNET_RESULTS>  -vv  
 
 
 
@@ -87,8 +87,8 @@ To evaluate the models run the following command.
 .. code-block:: python
 
 	./bin/scoring.py -df \
-	<PATH_TO_SCORES>/scores-dev -ef \
-	<PATH_TO_SCORES>/scores-eval
+	<FOLDER_TO_SAVE_FASNET_RESULTS>/grandtest/scores/scores-dev -ef \
+	<FOLDER_TO_SAVE_FASNET_RESULTS>/grandtest/scores/scores-eval
 
 
 

doc/running_mccn.md

@@ -100,8 +100,8 @@ To evaluate the models run the following command.
 .. code-block:: python
 
 	./bin/scoring.py -df \
-	<PATH_TO_SCORES>/scores-dev -ef \
-	<PATH_TO_SCORES>/scores-eval
+	<FOLDER_TO_SAVE_MCCNN_RESULTS>/grandtest/scores/scores-dev -ef \
+	<FOLDER_TO_SAVE_MCCNN_RESULTS>/grandtest/scores/scores-eval
 
 Using pretrained models
 =======================

setup.py

@@ -109,7 +109,7 @@ setup(
         'lbp-lr-infrared = bob.paper.mccnn.tifs2018.config.lbp_lr_infrared',
         'lbp-lr-thermal= bob.paper.mccnn.tifs2018.config.lbp_lr_thermal',
         'lbp-lr-depth= bob.paper.mccnn.tifs2018.config.lbp_lr_depth',
-        'fastnet = bob.paper.mccnn.tifs2018.config.FASNet_config',
+        'fasnet = bob.paper.mccnn.tifs2018.config.FASNet_config',
         'mccnn = bob.paper.mccnn.tifs2018.config.MCCNN_config',
         'iqm-lr = bob.paper.mccnn.tifs2018.config.iqm_lr',
         'haralick-svm = bob.paper.mccnn.tifs2018.config.haralick_svm',],