Refactor models, clean the code and separate training from testing.

25 files changed, 475 insertions, 469 deletions

diff --git a/config/dest_simple_mlp_2_cs.py b/config/dest_simple_mlp_2_cs.py
index accb611..be93427 100644
--- a/config/dest_simple_mlp_2_cs.py
+++ b/config/dest_simple_mlp_2_cs.py
@@ -1,8 +1,8 @@
 from blocks.initialization import IsotropicGaussian, Constant
 
-import model.dest_simple_mlp as model
-
 import data
+from model.dest_simple_mlp import Model, Stream
+
 
 n_begin_end_pts = 5     # how many points we consider at the beginning and end of the known trajectory
 n_end_pts = 5
diff --git a/config/dest_simple_mlp_2_cswdt.py b/config/dest_simple_mlp_2_cswdt.py
index 62d0db4..a86977c 100644
--- a/config/dest_simple_mlp_2_cswdt.py
+++ b/config/dest_simple_mlp_2_cswdt.py
@@ -1,8 +1,8 @@
-import model.dest_simple_mlp as model
-
 from blocks.initialization import IsotropicGaussian, Constant
 
 import data
+from model.dest_simple_mlp import Model, Stream
+
 
 n_begin_end_pts = 5     # how many points we consider at the beginning and end of the known trajectory
 n_end_pts = 5
diff --git a/config/dest_simple_mlp_2_noembed.py b/config/dest_simple_mlp_2_noembed.py
index bbe7798..5615a5c 100644
--- a/config/dest_simple_mlp_2_noembed.py
+++ b/config/dest_simple_mlp_2_noembed.py
@@ -1,8 +1,8 @@
-import model.dest_simple_mlp as model
-
 from blocks.initialization import IsotropicGaussian, Constant
 
 import data
+from model.dest_simple_mlp import Model, Stream
+
 
 n_begin_end_pts = 5     # how many points we consider at the beginning and end of the known trajectory
 n_end_pts = 5
diff --git a/config/dest_simple_mlp_tgtcls_0_cs.py b/config/dest_simple_mlp_tgtcls_0_cs.py
index 704e62c..f5b9a1e 100644
--- a/config/dest_simple_mlp_tgtcls_0_cs.py
+++ b/config/dest_simple_mlp_tgtcls_0_cs.py
@@ -3,8 +3,8 @@ import cPickle
 from blocks.initialization import IsotropicGaussian, Constant
 
 import data
+from model.dest_simple_mlp_tgtcls import Model, Stream
 
-import model.dest_simple_mlp_tgtcls as model
 
 n_begin_end_pts = 5     # how many points we consider at the beginning and end of the known trajectory
 n_end_pts = 5
diff --git a/config/dest_simple_mlp_tgtcls_1_cs.py b/config/dest_simple_mlp_tgtcls_1_cs.py
index f2a22a5..54457e7 100644
--- a/config/dest_simple_mlp_tgtcls_1_cs.py
+++ b/config/dest_simple_mlp_tgtcls_1_cs.py
@@ -3,8 +3,8 @@ import cPickle
 from blocks.initialization import IsotropicGaussian, Constant
 
 import data
+from model.dest_simple_mlp_tgtcls import Model, Stream
 
-import model.dest_simple_mlp_tgtcls as model
 
 n_begin_end_pts = 5     # how many points we consider at the beginning and end of the known trajectory
 n_end_pts = 5
diff --git a/config/dest_simple_mlp_tgtcls_1_cswdt.py b/config/dest_simple_mlp_tgtcls_1_cswdt.py
index a3ae654..486b7a6 100644
--- a/config/dest_simple_mlp_tgtcls_1_cswdt.py
+++ b/config/dest_simple_mlp_tgtcls_1_cswdt.py
@@ -3,8 +3,8 @@ import cPickle
 from blocks.initialization import IsotropicGaussian, Constant
 
 import data
+from model.dest_simple_mlp_tgtcls import Model, Stream
 
-import model.dest_simple_mlp_tgtcls as model
 
 n_begin_end_pts = 5     # how many points we consider at the beginning and end of the known trajectory
 n_end_pts = 5
diff --git a/config/dest_simple_mlp_tgtcls_1_cswdtx.py b/config/dest_simple_mlp_tgtcls_1_cswdtx.py
index 6306c15..713581d 100644
--- a/config/dest_simple_mlp_tgtcls_1_cswdtx.py
+++ b/config/dest_simple_mlp_tgtcls_1_cswdtx.py
@@ -3,8 +3,8 @@ import cPickle
 from blocks.initialization import IsotropicGaussian, Constant
 
 import data
+from model.dest_simple_mlp_tgtcls import Model, Stream
 
-import model.dest_simple_mlp_tgtcls as model
 
 n_begin_end_pts = 5     # how many points we consider at the beginning and end of the known trajectory
 n_end_pts = 5
diff --git a/config/dest_simple_mlp_tgtcls_1_cswdtx_alexandre.py b/config/dest_simple_mlp_tgtcls_1_cswdtx_alexandre.py
index 8c090c7..c0ba36c 100644
--- a/config/dest_simple_mlp_tgtcls_1_cswdtx_alexandre.py
+++ b/config/dest_simple_mlp_tgtcls_1_cswdtx_alexandre.py
@@ -3,8 +3,8 @@ import cPickle
 from blocks.initialization import IsotropicGaussian, Constant
 
 import data
+from model.dest_simple_mlp_tgtcls import Model, Stream
 
-import model.dest_simple_mlp_tgtcls as model
 
 n_begin_end_pts = 5     # how many points we consider at the beginning and end of the known trajectory
 n_end_pts = 5
diff --git a/config/joint_simple_mlp_tgtcls_111_cswdtx.py b/config/joint_simple_mlp_tgtcls_111_cswdtx.py
index deb6eba..a7e0415 100644
--- a/config/joint_simple_mlp_tgtcls_111_cswdtx.py
+++ b/config/joint_simple_mlp_tgtcls_111_cswdtx.py
@@ -1,10 +1,10 @@
 import cPickle
 
-import model.joint_simple_mlp_tgtcls as model
-
 from blocks.initialization import IsotropicGaussian, Constant
 
 import data
+from model.joint_simple_mlp_tgtcls import Model, Stream
+
 
 n_begin_end_pts = 5     # how many points we consider at the beginning and end of the known trajectory
 n_end_pts = 5
diff --git a/config/joint_simple_mlp_tgtcls_111_cswdtx_noise_dout.py b/config/joint_simple_mlp_tgtcls_111_cswdtx_noise_dout.py
new file mode 100644
index 0000000..1faea15
--- /dev/null
+++ b/config/joint_simple_mlp_tgtcls_111_cswdtx_noise_dout.py
@@ -0,0 +1,62 @@
+import cPickle
+
+from blocks import roles
+from blocks.bricks import Rectifier
+from blocks.filter import VariableFilter
+from blocks.initialization import IsotropicGaussian, Constant
+
+import data
+from model.joint_simple_mlp_tgtcls import Model, Stream
+
+
+n_begin_end_pts = 5     # how many points we consider at the beginning and end of the known trajectory
+n_end_pts = 5
+
+n_valid = 1000
+
+with open("%s/arrival-clusters.pkl" % data.path) as f:
+    dest_tgtcls = cPickle.load(f)
+
+# generate target classes for time prediction as a Fibonacci sequence
+time_tgtcls = [1, 2]
+for i in range(22):
+    time_tgtcls.append(time_tgtcls[-1] + time_tgtcls[-2])
+
+dim_embeddings = [
+    ('origin_call', data.origin_call_size+1, 10),
+    ('origin_stand', data.stands_size+1, 10),
+    ('week_of_year', 52, 10),
+    ('day_of_week', 7, 10),
+    ('qhour_of_day', 24 * 4, 10),
+    ('day_type', 3, 10),
+    ('taxi_id', 448, 10),
+]
+
+# Common network part
+dim_input = n_begin_end_pts * 2 * 2 + sum(x for (_, _, x) in dim_embeddings)
+dim_hidden = [500]
+
+# Destination prediction part
+dim_hidden_dest = [100]
+dim_output_dest = len(dest_tgtcls)
+
+# Time prediction part
+dim_hidden_time = [100]
+dim_output_time = len(time_tgtcls)
+
+# Cost ratio between distance cost and time cost
+time_cost_factor = 4
+
+embed_weights_init = IsotropicGaussian(0.001)
+mlp_weights_init = IsotropicGaussian(0.01)
+mlp_biases_init = Constant(0.001)
+
+batch_size = 200
+
+dropout = 0.5
+dropout_inputs = VariableFilter(bricks=[Rectifier], name='output')
+
+noise = 0.01
+noise_inputs = VariableFilter(roles=[roles.PARAMETER])
+
+valid_set = 'cuts/test_times_0'
diff --git a/config/joint_simple_mlp_tgtcls_1_cswdtx.py b/config/joint_simple_mlp_tgtcls_1_cswdtx.py
index 74b3c75..c4bc2fb 100644
--- a/config/joint_simple_mlp_tgtcls_1_cswdtx.py
+++ b/config/joint_simple_mlp_tgtcls_1_cswdtx.py
@@ -1,10 +1,10 @@
 import cPickle
 
-import model.joint_simple_mlp_tgtcls as model
-
 from blocks.initialization import IsotropicGaussian, Constant
 
 import data
+from model.joint_simple_mlp_tgtcls import Model, Stream
+
 
 n_begin_end_pts = 5     # how many points we consider at the beginning and end of the known trajectory
 n_end_pts = 5
diff --git a/config/time_simple_mlp_1.py b/config/time_simple_mlp_1.py
index bf3699d..c9203c8 100644
--- a/config/time_simple_mlp_1.py
+++ b/config/time_simple_mlp_1.py
@@ -1,8 +1,8 @@
-import model.time_simple_mlp as model
-
 from blocks.initialization import IsotropicGaussian, Constant
 
 import data
+from model.time_simple_mlp import Model, Stream
+
 
 n_begin_end_pts = 5     # how many points we consider at the beginning and end of the known trajectory
 n_end_pts = 5
diff --git a/config/time_simple_mlp_2_cswdtx.py b/config/time_simple_mlp_2_cswdtx.py
index 98467e3..dac585f 100644
--- a/config/time_simple_mlp_2_cswdtx.py
+++ b/config/time_simple_mlp_2_cswdtx.py
@@ -1,8 +1,8 @@
-import model.time_simple_mlp as model
-
 from blocks.initialization import IsotropicGaussian, Constant
 
 import data
+from model.time_simple_mlp import Model, Stream
+
 
 n_begin_end_pts = 5     # how many points we consider at the beginning and end of the known trajectory
 n_end_pts = 5
diff --git a/config/time_simple_mlp_tgtcls_2_cswdtx.py b/config/time_simple_mlp_tgtcls_2_cswdtx.py
index eb69714..eb06334 100644
--- a/config/time_simple_mlp_tgtcls_2_cswdtx.py
+++ b/config/time_simple_mlp_tgtcls_2_cswdtx.py
@@ -1,8 +1,8 @@
-import model.time_simple_mlp_tgtcls as model
-
 from blocks.initialization import IsotropicGaussian, Constant
 
 import data
+from model.time_simple_mlp_tgtcls import Model, Stream
+
 
 n_begin_end_pts = 5     # how many points we consider at the beginning and end of the known trajectory
 n_end_pts = 5
diff --git a/data/transformers.py b/data/transformers.py
index 57747fc..1b82dae 100644
--- a/data/transformers.py
+++ b/data/transformers.py
@@ -64,8 +64,9 @@ class TaxiGenerateSplits(Transformer):
 
         dlat = numpy.float32(self.data[self.id_latitude][-1])
         dlon = numpy.float32(self.data[self.id_longitude][-1])
+        ttime = numpy.int32(15 * (len(self.data[self.id_longitude]) - 1))
 
-        return tuple(r + [dlat, dlon, 15 * (len(self.data[self.id_longitude]) - 1)])
+        return tuple(r + [dlat, dlon, ttime])
 
 class TaxiAddFirstLastLen(Transformer):
     def __init__(self, k, stream):
@@ -87,7 +88,7 @@ class TaxiAddFirstLastLen(Transformer):
                             dtype=theano.config.floatX),
                   numpy.array(at_least_k(self.k, data[self.id_longitude], True, True)[-self.k:],
                               dtype=theano.config.floatX))
-        input_time = (15 * (len(data[self.id_latitude]) - 1),)
+        input_time = (numpy.int32(15 * (len(data[self.id_latitude]) - 1)),)
         return data + first_k + last_k + input_time
 
 class TaxiAddDateTime(Transformer):
@@ -101,7 +102,9 @@ class TaxiAddDateTime(Transformer):
         ts = data[self.id_timestamp]
         date = datetime.datetime.utcfromtimestamp(ts)
         yearweek = date.isocalendar()[1] - 1
-        info = ((51 if yearweek == 52 else yearweek), date.weekday(), date.hour * 4 + date.minute / 15)
+        info = (numpy.int8(51 if yearweek == 52 else yearweek),
+                numpy.int8(date.weekday()),
+                numpy.int8(date.hour * 4 + date.minute / 15))
         return data + info
 
 class TaxiExcludeTrips(Transformer):
diff --git a/model/__init__.py b/model/__init__.py
index e69de29..5c051f4 100644
--- a/model/__init__.py
+++ b/model/__init__.py
@@ -0,0 +1,36 @@
+from blocks.bricks import application, Initializable
+from blocks.bricks.lookup import LookupTable
+
+
+class ContextEmbedder(Initializable):
+    def __init__(self, config, **kwargs):
+        super(ContextEmbedder, self).__init__(**kwargs)
+        self.dim_embeddings = config.dim_embeddings
+        self.embed_weights_init = config.embed_weights_init
+
+        self.inputs = [ name for (name, _, _) in self.dim_embeddings ]
+        self.outputs = [ '%s_embedded' % name for name in self.inputs ]
+
+        self.lookups = { name: LookupTable(name='%s_lookup' % name) for name in self.inputs }
+        self.children = self.lookups.values()
+
+    def _push_allocation_config(self):
+        for (name, num, dim) in self.dim_embeddings:
+            self.lookups[name].length = num
+            self.lookups[name].dim = dim
+
+    def _push_initialization_config(self):
+        for name in self.inputs:
+            self.lookups[name].weights_init = self.embed_weights_init
+
+    @application
+    def apply(self, **kwargs):
+        return tuple(self.lookups[name].apply(kwargs[name]) for name in self.inputs)
+
+    @apply.property('inputs')
+    def apply_inputs(self):
+        return self.inputs
+
+    @apply.property('outputs')
+    def apply_outputs(self):
+        return self.outputs
diff --git a/model/dest_simple_mlp.py b/model/dest_simple_mlp.py
index a9e97cb..78d7131 100644
--- a/model/dest_simple_mlp.py
+++ b/model/dest_simple_mlp.py
@@ -1,71 +1,32 @@
-from blocks.bricks import MLP, Rectifier, Linear, Sigmoid, Identity
-from blocks.bricks.lookup import LookupTable
-
 from theano import tensor
+from blocks.bricks import application, Identity
 
 import data
 import error
+from model.mlp import FFMLP, Stream
 
-class Model(object):
-    def __init__(self, config):
-        # The input and the targets
-        x_firstk_latitude = (tensor.matrix('first_k_latitude') - data.train_gps_mean[0]) / data.train_gps_std[0]
-        x_firstk_longitude = (tensor.matrix('first_k_longitude') - data.train_gps_mean[1]) / data.train_gps_std[1]
-
-        x_lastk_latitude = (tensor.matrix('last_k_latitude') - data.train_gps_mean[0]) / data.train_gps_std[0]
-        x_lastk_longitude = (tensor.matrix('last_k_longitude') - data.train_gps_mean[1]) / data.train_gps_std[1]
-
-        input_list = [x_firstk_latitude, x_firstk_longitude, x_lastk_latitude, x_lastk_longitude]
-        embed_tables = []
-
-        self.require_inputs = ['first_k_latitude', 'first_k_longitude', 'last_k_latitude', 'last_k_longitude']
-
-        for (varname, num, dim) in config.dim_embeddings:
-            self.require_inputs.append(varname)
-            vardata = tensor.lvector(varname)
-            tbl = LookupTable(length=num, dim=dim, name='%s_lookup'%varname)
-            embed_tables.append(tbl)
-            input_list.append(tbl.apply(vardata))
-
-        y = tensor.concatenate((tensor.vector('destination_latitude')[:, None],
-                                tensor.vector('destination_longitude')[:, None]), axis=1)
-
-        # Define the model
-        mlp = MLP(activations=[Rectifier() for _ in config.dim_hidden] + [Identity()],
-                           dims=[config.dim_input] + config.dim_hidden + [config.dim_output])
-
-        # Create the Theano variables
-        inputs = tensor.concatenate(input_list, axis=1)
-        # inputs = theano.printing.Print("inputs")(inputs)
-        outputs = mlp.apply(inputs)
-
-        # Normalize & Center
-        # outputs = theano.printing.Print("normal_outputs")(outputs)
-        outputs = data.train_gps_std * outputs + data.train_gps_mean
-
-        # outputs = theano.printing.Print("outputs")(outputs)
-        # y = theano.printing.Print("y")(y)
 
-        outputs.name = 'outputs'
+class Model(FFMLP):
+    def __init__(self, config, **kwargs):
+        super(Model, self).__init__(config, output_layer=Identity, **kwargs)
 
-        # Calculate the cost
-        cost = error.erdist(outputs, y).mean()
-        cost.name = 'cost'
-        hcost = error.hdist(outputs, y).mean()
-        hcost.name = 'hcost'
+    @application(outputs=['destination'])
+    def predict(self, **kwargs):
+        outputs = super(Model, self).predict(**kwargs)
+        return data.train_gps_std * outputs + data.train_gps_mean
 
-        # Initialization
-        for tbl in embed_tables:
-            tbl.weights_init = config.embed_weights_init
-        mlp.weights_init = config.mlp_weights_init
-        mlp.biases_init = config.mlp_biases_init
+    @predict.property('inputs')
+    def predict_inputs(self):
+        return self.inputs
 
-        for tbl in embed_tables:
-            tbl.initialize()
-        mlp.initialize()
+    @application(outputs=['cost'])
+    def cost(self, **kwargs):
+        y_hat = self.predict(**kwargs)
+        y = tensor.concatenate((kwargs['destination_latitude'][:, None],
+                                kwargs['destination_longitude'][:, None]), axis=1)
 
-        self.cost = cost
-        self.monitor = [cost, hcost]
-        self.outputs = outputs
-        self.pred_vars = ['destination_latitude', 'destination_longitude']
+        return error.erdist(y_hat, y).mean()
 
+    @cost.property('inputs')
+    def cost_inputs(self):
+        return self.inputs + ['destination_latitude', 'destination_longitude']
diff --git a/model/dest_simple_mlp_tgtcls.py b/model/dest_simple_mlp_tgtcls.py
index 1381d4c..2d65097 100644
--- a/model/dest_simple_mlp_tgtcls.py
+++ b/model/dest_simple_mlp_tgtcls.py
@@ -1,73 +1,34 @@
-import numpy 
-
+import numpy
 import theano
 from theano import tensor
+from blocks.bricks import application, Softmax
 
-from blocks.bricks import MLP, Rectifier, Linear, Sigmoid, Identity, Softmax
-from blocks.bricks.lookup import LookupTable
-
-import data
 import error
+from model.mlp import FFMLP, Stream
 
-class Model(object):
-    def __init__(self, config):
-        # The input and the targets
-        x_firstk_latitude = (tensor.matrix('first_k_latitude') - data.train_gps_mean[0]) / data.train_gps_std[0]
-        x_firstk_longitude = (tensor.matrix('first_k_longitude') - data.train_gps_mean[1]) / data.train_gps_std[1]
-
-        x_lastk_latitude = (tensor.matrix('last_k_latitude') - data.train_gps_mean[0]) / data.train_gps_std[0]
-        x_lastk_longitude = (tensor.matrix('last_k_longitude') - data.train_gps_mean[1]) / data.train_gps_std[1]
-
-        input_list = [x_firstk_latitude, x_firstk_longitude, x_lastk_latitude, x_lastk_longitude]
-        embed_tables = []
-
-        self.require_inputs = ['first_k_latitude', 'first_k_longitude', 'last_k_latitude', 'last_k_longitude']
-
-        for (varname, num, dim) in config.dim_embeddings:
-            self.require_inputs.append(varname)
-            vardata = tensor.lvector(varname)
-            tbl = LookupTable(length=num, dim=dim, name='%s_lookup'%varname)
-            embed_tables.append(tbl)
-            input_list.append(tbl.apply(vardata))
-
-        y = tensor.concatenate((tensor.vector('destination_latitude')[:, None],
-                                tensor.vector('destination_longitude')[:, None]), axis=1)
-
-        # Define the model
-        mlp = MLP(activations=[Rectifier() for _ in config.dim_hidden] + [Softmax()],
-                           dims=[config.dim_input] + config.dim_hidden + [config.dim_output])
-        classes = theano.shared(numpy.array(config.tgtcls, dtype=theano.config.floatX), name='classes')
-
-        # Create the Theano variables
-        inputs = tensor.concatenate(input_list, axis=1)
-
-        # inputs = theano.printing.Print("inputs")(inputs)
-        cls_probas = mlp.apply(inputs)
-        outputs = tensor.dot(cls_probas, classes)
-
-        # outputs = theano.printing.Print("outputs")(outputs)
-        # y = theano.printing.Print("y")(y)
 
-        outputs.name = 'outputs'
+class Model(FFMLP):
+    def __init__(self, config, **kwargs):
+        super(Model, self, output_layer=Softmax).__init__(config, **kwargs)
+        self.classes = theano.shared(numpy.array(config.tgtcls, dtype=theano.config.floatX), name='classes')
 
-        # Calculate the cost
-        cost = error.erdist(outputs, y).mean()
-        cost.name = 'cost'
-        hcost = error.hdist(outputs, y).mean()
-        hcost.name = 'hcost'
+    @application(outputs=['destination'])
+    def predict(self, **kwargs):
+        cls_probas = super(Model, self).predict(**kwargs)
+        return tensor.dot(cls_probas, self.classes)
 
-        # Initialization
-        for tbl in embed_tables:
-            tbl.weights_init = config.embed_weights_init
-        mlp.weights_init = config.mlp_weights_init
-        mlp.biases_init = config.mlp_biases_init
+    @predict.property('inputs')
+    def predict_inputs(self):
+        return self.inputs
 
-        for tbl in embed_tables:
-            tbl.initialize()
-        mlp.initialize()
+    @application(outputs=['cost'])
+    def cost(self, **kwargs):
+        y_hat = self.predict(**kwargs)
+        y = tensor.concatenate((kwargs['destination_latitude'][:, None],
+                                kwargs['destination_longitude'][:, None]), axis=1)
 
-        self.cost = cost
-        self.monitor = [cost, hcost]
-        self.outputs = outputs
-        self.pred_vars = ['destination_latitude', 'destination_longitude']
+        return error.erdist(y_hat, y).mean()
 
+    @cost.property('inputs')
+    def cost_inputs(self):
+        return self.inputs + ['destination_latitude', 'destination_longitude']
diff --git a/model/joint_simple_mlp_tgtcls.py b/model/joint_simple_mlp_tgtcls.py
index 834afbf..d6d4e49 100644
--- a/model/joint_simple_mlp_tgtcls.py
+++ b/model/joint_simple_mlp_tgtcls.py
@@ -1,109 +1,71 @@
-from blocks.bricks import MLP, Rectifier, Linear, Sigmoid, Identity, Softmax
-from blocks.bricks.lookup import LookupTable
-
-from blocks.filter import VariableFilter
-from blocks.graph import ComputationGraph, apply_dropout
-
 import numpy
 import theano
 from theano import tensor
+from blocks import roles
+from blocks.bricks import application, MLP, Rectifier, Softmax
 
-import data
 import error
+from model.mlp import FFMLP, Stream
 
-class Model(object):
-    def __init__(self, config):
-        # The input and the targets
-        x_firstk_latitude = (tensor.matrix('first_k_latitude') - data.train_gps_mean[0]) / data.train_gps_std[0]
-        x_firstk_longitude = (tensor.matrix('first_k_longitude') - data.train_gps_mean[1]) / data.train_gps_std[1]
 
-        x_lastk_latitude = (tensor.matrix('last_k_latitude') - data.train_gps_mean[0]) / data.train_gps_std[0]
-        x_lastk_longitude = (tensor.matrix('last_k_longitude') - data.train_gps_mean[1]) / data.train_gps_std[1]
+class Model(FFMLP):
+    def __init__(self, config, **kwargs):
+        super(Model, self).__init__(config, **kwargs)
+        
+        self.dest_mlp = MLP(activations=[Rectifier() for _ in config.dim_hidden_dest] + [Softmax()],
+                       dims=[config.dim_hidden[-1]] + config.dim_hidden_dest + [config.dim_output_dest],
+                       name='dest_mlp')
+        self.time_mlp = MLP(activations=[Rectifier() for _ in config.dim_hidden_time] + [Softmax()],
+                       dims=[config.dim_hidden[-1]] + config.dim_hidden_time + [config.dim_output_time],
+                       name='time_mlp')
 
-        x_input_time = tensor.lvector('input_time')
+        self.dest_classes = theano.shared(numpy.array(config.dest_tgtcls, dtype=theano.config.floatX), name='dest_classes')
+        self.time_classes = theano.shared(numpy.array(config.time_tgtcls, dtype=theano.config.floatX), name='time_classes')
 
-        input_list = [x_firstk_latitude, x_firstk_longitude, x_lastk_latitude, x_lastk_longitude]
-        embed_tables = []
+        self.inputs.append('input_time')
+        self.children.extend([self.dest_mlp, self.time_mlp])
 
-        self.require_inputs = ['first_k_latitude', 'first_k_longitude', 'last_k_latitude', 'last_k_longitude', 'input_time']
+    def _push_initialization_config(self):
+        super(Model, self)._push_initialization_config()
+        for mlp in [self.dest_mlp, self.time_mlp]:
+            mlp.weights_init = self.config.mlp_weights_init
+            mlp.biases_init = self.config.mlp_biases_init
 
-        for (varname, num, dim) in config.dim_embeddings:
-            self.require_inputs.append(varname)
-            vardata = tensor.lvector(varname)
-            tbl = LookupTable(length=num, dim=dim, name='%s_lookup'%varname)
-            embed_tables.append(tbl)
-            input_list.append(tbl.apply(vardata))
+    @application(outputs=['destination', 'duration'])
+    def predict(self, **kwargs):
+        hidden = super(Model, self).predict(**kwargs)
 
-        y_dest = tensor.concatenate((tensor.vector('destination_latitude')[:, None],
-                                     tensor.vector('destination_longitude')[:, None]), axis=1)
-        y_time = tensor.lvector('travel_time')
+        dest_cls_probas = self.dest_mlp.apply(hidden)
+        dest_outputs = tensor.dot(dest_cls_probas, self.dest_classes)
 
-        # Define the model
-        common_mlp = MLP(activations=[Rectifier() for _ in config.dim_hidden],
-                           dims=[config.dim_input] + config.dim_hidden)
+        time_cls_probas = self.time_mlp.apply(hidden)
+        time_outputs = kwargs['input_time'] + tensor.dot(time_cls_probas, self.time_classes)
 
-        dest_mlp = MLP(activations=[Rectifier() for _ in config.dim_hidden_dest] + [Softmax()],
-                       dims=[config.dim_hidden[-1]] + config.dim_hidden_dest + [config.dim_output_dest],
-                       name='dest_mlp')
-        dest_classes = theano.shared(numpy.array(config.dest_tgtcls, dtype=theano.config.floatX), name='dest_classes')
+        self.add_auxiliary_variable(dest_cls_probas, name='destination classes ponderations')
+        self.add_auxiliary_variable(time_cls_probas, name='time classes ponderations')
 
-        time_mlp = MLP(activations=[Rectifier() for _ in config.dim_hidden_time] + [Softmax()],
-                       dims=[config.dim_hidden[-1]] + config.dim_hidden_time + [config.dim_output_time],
-                       name='time_mlp')
-        time_classes = theano.shared(numpy.array(config.time_tgtcls, dtype=theano.config.floatX), name='time_classes')
-
-        # Create the Theano variables
-        inputs = tensor.concatenate(input_list, axis=1)
-        # inputs = theano.printing.Print("inputs")(inputs)
-        hidden = common_mlp.apply(inputs)
-
-        dest_cls_probas = dest_mlp.apply(hidden)
-        dest_outputs = tensor.dot(dest_cls_probas, dest_classes)
-        dest_outputs.name = 'dest_outputs'
-
-        time_cls_probas = time_mlp.apply(hidden)
-        time_outputs = tensor.dot(time_cls_probas, time_classes) + x_input_time
-        time_outputs.name = 'time_outputs'
-
-        # Calculate the cost
-        dest_cost = error.erdist(dest_outputs, y_dest).mean()
-        dest_cost.name = 'dest_cost'
-        dest_hcost = error.hdist(dest_outputs, y_dest).mean()
-        dest_hcost.name = 'dest_hcost'
-
-        time_cost = error.rmsle(time_outputs.flatten(), y_time.flatten())
-        time_cost.name = 'time_cost'
-        time_scost = config.time_cost_factor * time_cost
-        time_scost.name = 'time_scost'
-
-        cost = dest_cost + time_scost
-
-        if hasattr(config, 'dropout_p'):
-            cg = ComputationGraph(cost)
-            dropout_inputs = VariableFilter(
-                    bricks=[b for b in list(common_mlp.children) +
-                                       list(dest_mlp.children) +
-                                       list(time_mlp.children)
-                              if isinstance(b, Rectifier)],
-                    name='output')(cg)
-            cg = apply_dropout(cg, dropout_inputs, config.dropout_p)
-            cost = cg.outputs[0]
-
-        cost.name = 'cost'
-
-        # Initialization
-        for tbl in embed_tables:
-            tbl.weights_init = config.embed_weights_init
-            tbl.initialize()
-
-        for mlp in [common_mlp, dest_mlp, time_mlp]:
-            mlp.weights_init = config.mlp_weights_init
-            mlp.biases_init = config.mlp_biases_init
-            mlp.initialize()
-
-        self.cost = cost
-        self.monitor = [cost, dest_cost, dest_hcost, time_cost, time_scost]
-        self.outputs = tensor.concatenate([dest_outputs, time_outputs[:, None]], axis=1)
-        self.outputs.name = 'outputs'
-        self.pred_vars = ['destination_longitude', 'destination_latitude', 'travel_time']
+        return (dest_outputs, time_outputs)
+
+    @predict.property('inputs')
+    def predict_inputs(self):
+        return self.inputs
+
+    @application(outputs=['cost'])
+    def cost(self, **kwargs):
+        (destination_hat, time_hat) = self.predict(**kwargs)
+
+        destination = tensor.concatenate((kwargs['destination_latitude'][:, None],
+                                          kwargs['destination_longitude'][:, None]), axis=1)
+        time = kwargs['travel_time']
+
+        destination_cost = error.erdist(destination_hat, destination).mean()
+        time_cost = error.rmsle(time_hat.flatten(), time.flatten())
+
+        self.add_auxiliary_variable(destination_cost, [roles.COST], 'destination_cost')
+        self.add_auxiliary_variable(time_cost, [roles.COST], 'time_cost')
+
+        return destination_cost + self.config.time_cost_factor * time_cost
 
+    @cost.property('inputs')
+    def cost_inputs(self):
+        return self.inputs + ['destination_latitude', 'destination_longitude', 'travel_time']
diff --git a/model/mlp.py b/model/mlp.py
new file mode 100644
index 0000000..9c84ef9
--- /dev/null
+++ b/model/mlp.py
@@ -0,0 +1,103 @@
+from theano import tensor
+
+from fuel.transformers import Batch
+from fuel.streams import DataStream
+from fuel.schemes import ConstantScheme, ShuffledExampleScheme
+from blocks.bricks import application, MLP, Rectifier, Initializable
+
+import data
+from data import transformers
+from data.hdf5 import TaxiDataset, TaxiStream
+from model import ContextEmbedder
+
+
+class FFMLP(Initializable):
+    def __init__(self, config, output_layer=None, **kwargs):
+        super(FFMLP, self).__init__(**kwargs)
+        self.config = config
+
+        self.context_embedder = ContextEmbedder(config)
+
+        output_activation = [] if output_layer is None else [output_layer()]
+        output_dim = [] if output_layer is None else [config.dim_output]
+        self.mlp = MLP(activations=[Rectifier() for _ in config.dim_hidden] + output_activation,
+                       dims=[config.dim_input] + config.dim_hidden + output_dim)
+
+        self.extremities = {'%s_k_%s' % (side, ['latitude', 'longitude'][axis]): axis for side in ['first', 'last'] for axis in [0, 1]}
+        self.inputs = self.context_embedder.inputs + self.extremities.keys()
+        self.children = [ self.context_embedder, self.mlp ]
+
+    def _push_initialization_config(self):
+        self.mlp.weights_init = self.config.mlp_weights_init
+        self.mlp.biases_init = self.config.mlp_biases_init
+
+    @application(outputs=['prediction'])
+    def predict(self, **kwargs):
+        embeddings = tuple(self.context_embedder.apply(**{k: kwargs[k] for k in self.context_embedder.inputs }))
+        extremities = tuple((kwargs[k] - data.train_gps_mean[v]) / data.train_gps_std[v] for k, v in self.extremities.items())
+
+        inputs = tensor.concatenate(extremities + embeddings, axis=1)
+        outputs = self.mlp.apply(inputs)
+
+        return outputs
+
+    @predict.property('inputs')
+    def predict_inputs(self):
+        return self.inputs
+
+class Stream(object):
+    def __init__(self, config):
+        self.config = config
+
+    def train(self, req_vars):
+        stream = TaxiDataset('train')
+        stream = DataStream(stream, iteration_scheme=ShuffledExampleScheme(stream.num_examples))
+
+        valid = TaxiDataset(self.config.valid_set, 'valid.hdf5', sources=('trip_id',))
+        valid_trips_ids = valid.get_data(None, slice(0, valid.num_examples))[0]
+
+        stream = transformers.TaxiExcludeTrips(valid_trips_ids, stream)
+        stream = transformers.TaxiGenerateSplits(stream, max_splits=100)
+
+        stream = transformers.TaxiAddDateTime(stream)
+        stream = transformers.TaxiAddFirstLastLen(self.config.n_begin_end_pts, stream)
+        stream = transformers.Select(stream, tuple(req_vars))
+        return Batch(stream, iteration_scheme=ConstantScheme(self.config.batch_size))
+
+    def valid(self, req_vars):
+        stream = TaxiStream(self.config.valid_set, 'valid.hdf5')
+
+        stream = transformers.TaxiAddDateTime(stream)
+        stream = transformers.TaxiAddFirstLastLen(self.config.n_begin_end_pts, stream)
+        stream = transformers.Select(stream, tuple(req_vars))
+        return Batch(stream, iteration_scheme=ConstantScheme(1000))
+
+    def test(self, req_vars):
+        stream = TaxiStream('test')
+        
+        stream = transformers.TaxiAddDateTime(stream)
+        stream = transformers.TaxiAddFirstLastLen(self.config.n_begin_end_pts, stream)
+
+        return Batch(stream, iteration_scheme=ConstantScheme(1))
+
+    def inputs(self):
+        return {'call_type': tensor.bvector('call_type'),
+                'origin_call': tensor.ivector('origin_call'),
+                'origin_stand': tensor.bvector('origin_stand'),
+                'taxi_id': tensor.wvector('taxi_id'),
+                'timestamp': tensor.ivector('timestamp'),
+                'day_type': tensor.bvector('day_type'),
+                'missing_data': tensor.bvector('missing_data'),
+                'latitude': tensor.matrix('latitude'),
+                'longitude': tensor.matrix('longitude'),
+                'destination_latitude': tensor.vector('destination_latitude'),
+                'destination_longitude': tensor.vector('destination_longitude'),
+                'travel_time': tensor.ivector('travel_time'),
+                'first_k_latitude': tensor.matrix('first_k_latitude'),
+                'first_k_longitude': tensor.matrix('first_k_longitude'),
+                'last_k_latitude': tensor.matrix('last_k_latitude'),
+                'last_k_longitude': tensor.matrix('last_k_longitude'),
+                'input_time': tensor.ivector('input_time'),
+                'week_of_year': tensor.bvector('week_of_year'),
+                'day_of_week': tensor.bvector('day_of_week'),
+                'qhour_of_day': tensor.bvector('qhour_of_day')}
diff --git a/model/time_simple_mlp.py b/model/time_simple_mlp.py
index 9e1b10a..0ce6b29 100644
--- a/model/time_simple_mlp.py
+++ b/model/time_simple_mlp.py
@@ -1,63 +1,29 @@
-from blocks.bricks import MLP, Rectifier, Linear, Sigmoid, Identity
-from blocks.bricks.lookup import LookupTable
+from blocks.bricks import application, Identity
 
-from theano import tensor
-
-import data
 import error
+from model.mlp import FFMLP, Stream
 
-class Model(object):
-    def __init__(self, config):
-        # The input and the targets
-        x_firstk_latitude = (tensor.matrix('first_k_latitude') - data.train_gps_mean[0]) / data.train_gps_std[0]
-        x_firstk_longitude = (tensor.matrix('first_k_longitude') - data.train_gps_mean[1]) / data.train_gps_std[1]
-
-        x_lastk_latitude = (tensor.matrix('last_k_latitude') - data.train_gps_mean[0]) / data.train_gps_std[0]
-        x_lastk_longitude = (tensor.matrix('last_k_longitude') - data.train_gps_mean[1]) / data.train_gps_std[1]
-
-        input_list = [x_firstk_latitude, x_firstk_longitude, x_lastk_latitude, x_lastk_longitude]
-        embed_tables = []
-
-        self.require_inputs = ['first_k_latitude', 'first_k_longitude', 'last_k_latitude', 'last_k_longitude']
-
-        for (varname, num, dim) in config.dim_embeddings:
-            self.require_inputs.append(varname)
-            vardata = tensor.lvector(varname)
-            tbl = LookupTable(length=num, dim=dim, name='%s_lookup'%varname)
-            embed_tables.append(tbl)
-            input_list.append(tbl.apply(vardata))
-
-        y = tensor.lvector('travel_time')
-
-        # Define the model
-        mlp = MLP(activations=[Rectifier() for _ in config.dim_hidden] + [Identity()],
-                           dims=[config.dim_input] + config.dim_hidden + [config.dim_output])
-
-        # Create the Theano variables
-        inputs = tensor.concatenate(input_list, axis=1)
-        # inputs = theano.printing.Print("inputs")(inputs)
-        outputs = config.exp_base ** mlp.apply(inputs)
-
-        # outputs = theano.printing.Print("outputs")(outputs)
-        # y = theano.printing.Print("y")(y)
 
-        outputs.name = 'outputs'
+class Model(FFMLP):
+    def __init__(self, config, **kwargs):
+        super(Model, self).__init__(config, output_layer=Identity, **kwargs)
+        self.inputs.append('input_time')
 
-        # Calculate the cost
-        cost = error.rmsle(outputs.flatten(), y.flatten())
-        cost.name = 'cost'
+    @application(outputs=['duration'])
+    def predict(self, **kwargs):
+        outputs = super(Model, self).predict(**kwargs)
+        return kwargs['input_time'] + self.config.exp_base ** outputs
 
-        # Initialization
-        for tbl in embed_tables:
-            tbl.weights_init = config.embed_weights_init
-        mlp.weights_init = config.mlp_weights_init
-        mlp.biases_init = config.mlp_biases_init
+    @predict.property('inputs')
+    def predict_inputs(self):
+        return self.inputs
 
-        for tbl in embed_tables:
-            tbl.initialize()
-        mlp.initialize()
+    @application(outputs=['cost'])
+    def cost(self, **kwargs):
+        y_hat = self.predict(**kwargs)
+        y = kwargs['travel_time']
+        return error.rmsle(y_hat.flatten(), y.flatten())
 
-        self.cost = cost
-        self.monitor = [cost]
-        self.outputs = outputs
-        self.pred_vars = ['travel_time']
+    @cost.property('inputs')
+    def cost_inputs(self):
+        return self.inputs + ['travel_time']
diff --git a/model/time_simple_mlp_tgtcls.py b/model/time_simple_mlp_tgtcls.py
index 1f1eab7..35c8d8a 100644
--- a/model/time_simple_mlp_tgtcls.py
+++ b/model/time_simple_mlp_tgtcls.py
@@ -1,67 +1,33 @@
-from blocks.bricks import MLP, Rectifier, Linear, Sigmoid, Identity, Softmax
-from blocks.bricks.lookup import LookupTable
-
 import numpy
 import theano
 from theano import tensor
+from blocks.bricks import application, Softmax
 
-import data
 import error
+from model.mlp import FFMLP, Stream
 
-class Model(object):
-    def __init__(self, config):
-        # The input and the targets
-        x_firstk_latitude = (tensor.matrix('first_k_latitude') - data.train_gps_mean[0]) / data.train_gps_std[0]
-        x_firstk_longitude = (tensor.matrix('first_k_longitude') - data.train_gps_mean[1]) / data.train_gps_std[1]
-
-        x_lastk_latitude = (tensor.matrix('last_k_latitude') - data.train_gps_mean[0]) / data.train_gps_std[0]
-        x_lastk_longitude = (tensor.matrix('last_k_longitude') - data.train_gps_mean[1]) / data.train_gps_std[1]
-
-        input_list = [x_firstk_latitude, x_firstk_longitude, x_lastk_latitude, x_lastk_longitude]
-        embed_tables = []
-
-        self.require_inputs = ['first_k_latitude', 'first_k_longitude', 'last_k_latitude', 'last_k_longitude']
-
-        for (varname, num, dim) in config.dim_embeddings:
-            self.require_inputs.append(varname)
-            vardata = tensor.lvector(varname)
-            tbl = LookupTable(length=num, dim=dim, name='%s_lookup'%varname)
-            embed_tables.append(tbl)
-            input_list.append(tbl.apply(vardata))
-
-        y = tensor.lvector('travel_time')
-
-        # Define the model
-        mlp = MLP(activations=[Rectifier() for _ in config.dim_hidden] + [Softmax()],
-                           dims=[config.dim_input] + config.dim_hidden + [config.dim_output])
-        classes = theano.shared(numpy.array(config.tgtcls, dtype=theano.config.floatX), name='classes')
-
-        # Create the Theano variables
-        inputs = tensor.concatenate(input_list, axis=1)
-        # inputs = theano.printing.Print("inputs")(inputs)
-        cls_probas = mlp.apply(inputs)
-        outputs = tensor.dot(cls_probas, classes)
-
-        # outputs = theano.printing.Print("outputs")(outputs)
-        # y = theano.printing.Print("y")(y)
 
-        outputs.name = 'outputs'
+class Model(FFMLP):
+    def __init__(self, config, **kwargs):
+        super(Model, self, output_layer=Softmax).__init__(config, **kwargs)
+        self.classes = theano.shared(numpy.array(config.tgtcls, dtype=theano.config.floatX), name='classes')
+        self.inputs.append('input_time')
 
-        # Calculate the cost
-        cost = error.rmsle(outputs.flatten(), y.flatten())
-        cost.name = 'cost'
+    @application(outputs=['duration'])
+    def predict(self, **kwargs):
+        cls_probas = super(Model, self).predict(**kwargs)
+        return kwargs['input_time'] + tensor.dot(cls_probas, self.classes)
 
-        # Initialization
-        for tbl in embed_tables:
-            tbl.weights_init = config.embed_weights_init
-        mlp.weights_init = config.mlp_weights_init
-        mlp.biases_init = config.mlp_biases_init
+    @predict.property('inputs')
+    def predict_inputs(self):
+        return self.inputs
 
-        for tbl in embed_tables:
-            tbl.initialize()
-        mlp.initialize()
+    @application(outputs=['cost'])
+    def cost(self, **kwargs):
+        y_hat = self.predict(**kwargs)
+        y = kwargs['travel_time']
+        return error.rmsle(y_hat.flatten(), y.flatten())
 
-        self.cost = cost
-        self.monitor = [cost]
-        self.outputs = outputs
-        self.pred_vars = ['travel_time']
+    @cost.property('inputs')
+    def cost_inputs(self):
+        return self.inputs + ['travel_time']
diff --git a/test.py b/test.py
new file mode 100755
index 0000000..bf39b15
--- /dev/null
+++ b/test.py
@@ -0,0 +1,48 @@
+#!/usr/bin/env python
+
+import sys
+import os
+import importlib
+import csv
+
+from blocks.dump import load_parameter_values
+from blocks.model import Model
+
+
+if __name__ == "__main__":
+    if len(sys.argv) != 2:
+        print >> sys.stderr, 'Usage: %s config' % sys.argv[0]
+        sys.exit(1)
+    model_name = sys.argv[1]
+    config = importlib.import_module('.%s' % model_name, 'config')
+    model_config = config.Model(config)
+
+    stream = config.Stream(config)
+    inputs = stream.inputs()
+    outputs = model_config.predict.outputs
+    req_vars_test = model_config.predict.inputs + ['trip_id']
+    test_stream = stream.test(req_vars_test)
+
+    model = Model(model_config.predict(**inputs))
+    parameters = load_parameter_values(os.path.join('model_data', model_name, 'params.npz'))
+    model.set_param_values(parameters)
+
+    if 'destination' in outputs:
+        dest_outfile = open("output/test-dest-output-%s.csv" % model_name, "w")
+        dest_outcsv = csv.writer(dest_outfile)
+        dest_outcsv.writerow(["TRIP_ID", "LATITUDE", "LONGITUDE"])
+    if 'duration' in outputs:
+        time_outfile = open("output/test-time-output-%s.csv" % model_name, "w")
+        time_outcsv = csv.writer(time_outfile)
+        time_outcsv.writerow(["TRIP_ID", "TRAVEL_TIME"])
+
+    function = model.get_theano_function()
+    for d in test_stream.get_epoch_iterator(as_dict=True):
+        input_values = [d[k.name] for k in model.inputs]
+        output_values = function(*input_values)
+        if 'destination' in outputs:
+            destination = output_values[outputs.index('destination')]
+            dest_outcsv.writerow([d['trip_id'][0], destination[0, 0], destination[0, 1]])
+        if 'duration' in outputs:
+            duration = output_values[outputs.index('duration')]
+            time_outcsv.writerow([d['trip_id'][0], duration[0]])
diff --git a/train.py b/train.py
index 96dd798..97d8bb3 100755
--- a/train.py
+++ b/train.py
@@ -1,32 +1,25 @@
 #!/usr/bin/env python
 
-import sys
-import logging
 import importlib
+import logging
+import operator
+import os
+import sys
+from functools import reduce
 
-import csv
-
-from picklable_itertools.extras import equizip
-
-from blocks.model import Model
-
-from fuel.transformers import Batch
-from fuel.streams import DataStream
-from fuel.schemes import ConstantScheme, ShuffledExampleScheme
-
-from blocks.algorithms import CompositeRule, RemoveNotFinite, GradientDescent, AdaDelta, Momentum
-from blocks.graph import ComputationGraph, apply_dropout
-from blocks.main_loop import MainLoop
+from blocks import roles
+from blocks.algorithms import AdaDelta, CompositeRule, GradientDescent, RemoveNotFinite
 from blocks.extensions import Printing, FinishAfter
-from blocks.extensions.saveload import Dump, LoadFromDump, Checkpoint
 from blocks.extensions.monitoring import DataStreamMonitoring, TrainingDataMonitoring
 from blocks.extensions.plot import Plot
+from blocks.extensions.saveload import Dump, LoadFromDump
+from blocks.filter import VariableFilter
+from blocks.graph import ComputationGraph, apply_dropout, apply_noise
+from blocks.main_loop import MainLoop
+from blocks.model import Model
 
-from theano import tensor
 
-from data import transformers
-from data.hdf5 import TaxiDataset, TaxiStream
-import apply_model
+logger = logging.getLogger(__name__)
 
 if __name__ == "__main__":
     if len(sys.argv) != 2:
@@ -35,123 +28,70 @@ if __name__ == "__main__":
     model_name = sys.argv[1]
     config = importlib.import_module('.%s' % model_name, 'config')
 
-def compile_valid_trip_ids():
-    valid = TaxiDataset(config.valid_set, 'valid.hdf5', sources=('trip_id',))
-    ids = valid.get_data(None, slice(0, valid.num_examples))
-    return set(ids[0])
-
-def setup_train_stream(req_vars, valid_trips_ids):
-    train = TaxiDataset('train')
-    train = DataStream(train, iteration_scheme=ShuffledExampleScheme(train.num_examples))
-
-    train = transformers.TaxiExcludeTrips(valid_trips_ids, train)
-    train = transformers.TaxiGenerateSplits(train, max_splits=100)
-
-    train = transformers.TaxiAddDateTime(train)
-    train = transformers.TaxiAddFirstLastLen(config.n_begin_end_pts, train)
-    train = transformers.Select(train, tuple(req_vars))
-
-    train_stream = Batch(train, iteration_scheme=ConstantScheme(config.batch_size))
-
-    return train_stream
-
-def setup_valid_stream(req_vars):
-    valid = TaxiStream(config.valid_set, 'valid.hdf5')
+    logger.info('# Configuration: %s' % config.__name__)
+    for key in dir(config):
+        if not key.startswith('__') and isinstance(getattr(config, key), (int, str, list, tuple)):
+            logger.info('    %20s %s' % (key, str(getattr(config, key))))
 
-    valid = transformers.TaxiAddDateTime(valid)
-    valid = transformers.TaxiAddFirstLastLen(config.n_begin_end_pts, valid)
-    valid = transformers.Select(valid, tuple(req_vars))
+    model = config.Model(config)
+    model.initialize()
 
-    valid_stream = Batch(valid, iteration_scheme=ConstantScheme(1000))
-    
-    return valid_stream
-
-def setup_test_stream(req_vars):
-    test = TaxiStream('test')
-    
-    test = transformers.TaxiAddDateTime(test)
-    test = transformers.TaxiAddFirstLastLen(config.n_begin_end_pts, test)
-    test = transformers.Select(test, tuple(req_vars))
-
-    test_stream = Batch(test, iteration_scheme=ConstantScheme(1000))
-
-    return test_stream
-
-
-def main():
-    model = config.model.Model(config)
-
-    cost = model.cost
-    outputs = model.outputs
-
-    req_vars = model.require_inputs + model.pred_vars
-    req_vars_test = model.require_inputs + [ 'trip_id' ]
+    stream = config.Stream(config)
+    inputs = stream.inputs()
+    req_vars = model.cost.inputs
 
-    valid_trips_ids = compile_valid_trip_ids()
-    train_stream = setup_train_stream(req_vars, valid_trips_ids)
-    valid_stream = setup_valid_stream(req_vars)
+    train_stream = stream.train(req_vars)
+    valid_stream = stream.valid(req_vars)
 
-    # Training
+    cost = model.cost(**inputs)
     cg = ComputationGraph(cost)
+    unmonitor = set()
+    if hasattr(config, 'dropout') and config.dropout < 1.0:
+        unmonitor.update(VariableFilter(roles=[roles.COST])(cg.variables))
+        cg = apply_dropout(cg, config.dropout_inputs(cg), config.dropout)
+    if hasattr(config, 'noise') and config.noise > 0.0:
+        unmonitor.update(VariableFilter(roles=[roles.COST])(cg.variables))
+        cg = apply_noise(cg, config.noise_inputs(cg), config.noise)
+    cost = cg.outputs[0]
+    cg = Model(cost)
+
+    logger.info('# Parameter shapes:')
+    parameters_size = 0
+    for key, value in cg.get_params().iteritems():
+        logger.info('    %20s %s' % (value.get_value().shape, key))
+        parameters_size += reduce(operator.mul, value.get_value().shape, 1)
+    logger.info('Total number of parameters: %d in %d matrices' % (parameters_size, len(cg.get_params())))
 
     params = cg.parameters
-
     algorithm = GradientDescent(
         cost=cost,
         step_rule=CompositeRule([
                 RemoveNotFinite(),
-                #AdaDelta(decay_rate=0.95),
-                Momentum(learning_rate=config.learning_rate, momentum=config.momentum),
-        ]),
+                AdaDelta(),
+            ]),
         params=params)
     
-    plot_vars = [['valid_' + x.name for x in model.monitor]]
-    print "Plot: ", plot_vars
-
-    extensions=[TrainingDataMonitoring(model.monitor, prefix='train', every_n_batches=1000),
-                DataStreamMonitoring(model.monitor, valid_stream,
+    monitored = set([cost] + VariableFilter(roles=[roles.COST])(cg.variables)) - unmonitor
+    plot_vars = [['valid_' + x.name for x in monitored]]
+    logger.info('Plotted variables: %s' % str(plot_vars))
+
+    dump_path = os.path.join('model_data', model_name)
+    logger.info('Dump path: %s' % dump_path)
+    extensions=[TrainingDataMonitoring(monitored, prefix='train', every_n_batches=1000),
+                DataStreamMonitoring(monitored, valid_stream,
                                      prefix='valid',
-                                     every_n_batches=500),
-                Printing(every_n_batches=500),
+                                     every_n_batches=1000),
+                Printing(every_n_batches=1000),
                 Plot(model_name, channels=plot_vars, every_n_batches=500),
-                # Checkpoint('model.pkl', every_n_batches=100),
-                Dump('model_data/' + model_name, every_n_batches=500),
-                LoadFromDump('model_data/' + model_name),
-                # FinishAfter(after_epoch=4),
+                Dump(dump_path, every_n_batches=5000),
+                LoadFromDump(dump_path),
+                #FinishAfter(after_n_batches=2),
                 ]
 
     main_loop = MainLoop(
-        model=Model([cost]),
+        model=cg,
         data_stream=train_stream,
         algorithm=algorithm,
         extensions=extensions)
     main_loop.run()
     main_loop.profile.report()
-
-    # Produce an output on the test data
-    test_stream = setup_test_stream(req_vars_test)
-
-    if 'destination_longitude' in model.pred_vars:
-        dest_outfile = open("output/test-dest-output-%s.csv" % model_name, "w")
-        dest_outcsv = csv.writer(dest_outfile)
-        dest_outcsv.writerow(["TRIP_ID", "LATITUDE", "LONGITUDE"])
-    if 'travel_time' in model.pred_vars:
-        time_outfile = open("output/test-time-output-%s.csv" % model_name, "w")
-        time_outcsv = csv.writer(time_outfile)
-        time_outcsv.writerow(["TRIP_ID", "TRAVEL_TIME"])
-
-    for out in apply_model.Apply(outputs=outputs, stream=test_stream, return_vars=['trip_id', 'outputs']):
-        outputs = out['outputs']
-        for i, trip in enumerate(out['trip_id']):
-            if model.pred_vars == ['travel_time']:
-                time_outcsv.writerow([trip, int(outputs[i])])
-            else:
-                dest_outcsv.writerow([trip, repr(outputs[i, 0]), repr(outputs[i, 1])])
-                if 'travel_time' in model.pred_vars:
-                    time_outcsv.writerow([trip, int(outputs[i, 2])])
-
-
-if __name__ == "__main__":
-    logging.basicConfig(level=logging.INFO)
-    main()
-
diff --git a/visualizer/HTTPServer.py b/visualizer/HTTPServer.py
index e71bef3..52b2531 100755
--- a/visualizer/HTTPServer.py
+++ b/visualizer/HTTPServer.py
@@ -7,8 +7,6 @@ import SimpleHTTPServer
 import SocketServer
 from cStringIO import StringIO
 
-import h5py
-
 import data
 from data.hdf5 import TaxiDataset
 from visualizer import Vlist, Path