1 files changed, 18 insertions, 11 deletions
diff --git a/simulation/active_blocks.py b/simulation/active_blocks.py
index 40266b2..32ce927 100644
--- a/simulation/active_blocks.py
+++ b/simulation/active_blocks.py
@@ -27,10 +27,7 @@ class LearnedDataset(fuel.datasets.Dataset):
         self.source = lambda graph: utils.random_source(graph, self.node_p)
 
     def get_data(self, state=None, request=None):
-        # floatX = 'int8'
-        x_obs, s_obs = utils.simulate_cascades(request, self.graph, self.source)
-
-        return (x_obs, s_obs)
+        return utils.simulate_cascades(request, self.graph, self.source)
 
 
 class ActiveLearning(blocks.extensions.SimpleExtension):
@@ -87,15 +84,23 @@ def create_mle_model(graph):
 
 
 def rmse_error(graph, params):
-    n_nodes = len(graph)
-    g_shared = theano.shared(value=graph, name='graph')
-    diff = (g_shared - params) ** 2
-    subarray = tsr.arange(g_shared.shape[0])
+    n_nodes = graph.shape[0]
+    diff = (graph - params) ** 2
+    subarray = tsr.arange(n_nodes)
     tsr.set_subtensor(diff[subarray, subarray], 0)
     rmse = tsr.sum(diff) / (n_nodes ** 2)
     rmse.name = 'rmse'
-    g_shared.name = 'graph'
-    return rmse, g_shared
+    return rmse
+
+
+def relative_error(graph, params):
+    n_nodes = graph.shape[0]
+    diff = abs(g_shared - params)
+    subarray = tsr.arange(n_nodes)
+    tsr.set_subtensor(diff[subarray, subarray], 0)
+    error = tsr.sum(tsr.switch(tsr.eq(graph, 0.), 0., diff / graph)) / n_nodes
+    error.name = 'rel_error'
+    return error
 
 
 def create_fixed_data_stream(n_obs, graph, batch_size, shuffle=True):
@@ -128,8 +133,10 @@ if __name__ == "__main__":
     graph = utils.create_wheel(10)
     print('GRAPH:\n', graph, '\n-------------\n')
 
+    g_shared = theano.shared(value=graph, name='graph')
     x, s, params, cost = create_mle_model(graph)
-    rmse, g_shared = rmse_error(graph, params)
+    rmse = rmse_error(g_shared, params)
+    error = relative_error(g_shared, params)
 
     alg = algorithms.GradientDescent(
         cost=-cost, parameters=[params], step_rule=blocks.algorithms.AdaDelta()