Visual Debugging

code/experiment.py

@@ -67,7 +67,7 @@ class FixpointExperiment(Experiment):
             net.self_attack()
             i += 1
             if run_id:
-                weights = net.get_weights()
+                weights = net.get_weights_flat()
                 self.add_trajectory_segment(run_id, weights)
         self.count(net)
 

code/methods.py

@@ -173,6 +173,7 @@ class FeedForwardNetwork(_BaseNetwork):
                 bar.update()
         return losses
 
+
 if __name__ == '__main__':
     with Experiment() as exp:
         features, cells, layers = 2, 2, 2

code/network.py

@@ -83,6 +83,9 @@ class NeuralNetwork(PrintingObject):
     def get_weights(self):
         return self.model.get_weights()
 
+    def get_weights_flat(self):
+        return np.hstack([weight.flatten() for weight in self.get_weights()])
+
     def set_weights(self, new_weights):
         return self.model.set_weights(new_weights)
 
@@ -603,9 +606,9 @@ class TrainingNeuralNetworkDecorator(NeuralNetwork):
 
 
 if __name__ == '__main__':
-    if False:
+    if True:
         with FixpointExperiment() as exp:
-            for run_id in tqdm(range(1)):
+            for run_id in tqdm(range(100)):
                 # net = WeightwiseNeuralNetwork(width=2, depth=2).with_keras_params(activation='linear')
                 # net = AggregatingNeuralNetwork(aggregates=4, width=2, depth=2)\
                 net = FFTNeuralNetwork(aggregates=4, width=2, depth=2) \

code/visualization.py

@@ -22,44 +22,39 @@ def build_args():
 
 def plot_latent_trajectories(data_dict, filename='latent_trajectory_plot'):
 
-    # TODO Fist and Last Position Markers
-
-    def norm(val, a=0, b=0.25):
-        return (val - a) / (b - a)
-
     bupu = cl.scales['9']['seq']['BuPu']
-    scale = cl.interp(bupu, len(data_dict))  # Map color scale to N bins
+    scale = cl.interp(bupu, len(data_dict)+1)  # Map color scale to N bins
 
     # Fit the mebedding space
     transformer = TSNE()
-    for trajectory in data_dict:
+    for trajectory_id in data_dict:
-        transformer.fit(trajectory)
+        transformer.fit(np.asarray(data_dict[trajectory_id]))
 
     # Transform data accordingly and plot it
     data = []
-    for t_id, trajectory in enumerate(data_dict):
+    for trajectory_id in data_dict:
-        transformed = transformer.fit(trajectory)
+        transformed = transformer._fit(np.asarray(data_dict[trajectory_id]))
         line_trace = go.Scatter(
             x=transformed[:, 0],
             y=transformed[:, 1],
             text='Hovertext goes here'.format(),
-            line=dict(color=scale[t_id]),
+            line=dict(color=scale[trajectory_id]),
             # legendgroup='Position -{}'.format(pos),
             # name='Position -{}'.format(pos),
             showlegend=False,
             # hoverinfo='text',
             mode='lines')
-        line_start = go.Scatter(mode='markers', x=transformed[0, 0], y=transformed[0, 1],
+        line_start = go.Scatter(mode='markers', x=[transformed[0, 0]], y=[transformed[0, 1]],
                                 marker=dict(
                                     color='rgb(255, 0, 0)',
-                                    size=2
+                                    size=4
                                 ),
                                 showlegend=False
                                 )
-        line_end = go.Scatter(mode='markers', x=transformed[-1, 0], y=transformed[-1, 1],
+        line_end = go.Scatter(mode='markers', x=[transformed[-1, 0]], y=[transformed[-1, 1]],
                               marker=dict(
                                   color='rgb(0, 0, 0)',
-                                  size=2
+                                  size=4
                               ),
                               showlegend=False
                               )
@@ -74,30 +69,30 @@ def plot_latent_trajectories(data_dict, filename='latent_trajectory_plot'):
     pass
 
 
-def plot_latent_trajectories_3D(param_dict, filename='plot'):
+def plot_latent_trajectories_3D(data_dict, filename='plot'):
     def norm(val, a=0, b=0.25):
         return (val - a) / (b - a)
 
     bupu = cl.scales['9']['seq']['BuPu']
-    scale = cl.interp(bupu, len(param_dict.get('trajectories', [])))  # Map color scale to N bins
+    scale = cl.interp(bupu, len(data_dict)+1)  # Map color scale to N bins
 
-    max_len = max([len(trajectory) for trajectory in param_dict.get('trajectories', [])])
+    max_len = max([len(trajectory) for trajectory in data_dict.values()])
 
     # Fit the mebedding space
     transformer = TSNE()
-    for trajectory in param_dict.get('trajectories', []):
+    for trajectory_id in data_dict:
-        transformer.fit(trajectory)
+        transformer.fit(data_dict[trajectory_id])
 
     # Transform data accordingly and plot it
     data = []
-    for t_id, trajectory in enumerate(param_dict.get('trajectories', [])):
+    for trajectory_id in data_dict:
-        transformed = transformer.fit(trajectory)
+        transformed = transformer._fit(np.asarray(data_dict[trajectory_id]))
         trace = go.Scatter3d(
             x=transformed[:, 0],
             y=transformed[:, 1],
-            z=np.arange(max(max_len)),
+            z=np.arange(transformed.shape[0]),
             text='Hovertext goes here'.format(),
-            line=dict(color=scale[t_id]),
+            line=dict(color=scale[trajectory_id]),
             # legendgroup='Position -{}'.format(pos),
             # name='Position -{}'.format(pos),
             showlegend=False,
@@ -106,9 +101,9 @@ def plot_latent_trajectories_3D(param_dict, filename='plot'):
         data.append(trace)
 
     layout = go.Layout(scene=dict(aspectratio=dict(x=2, y=2, z=1),
-                         xaxis=dict(tickwidth=1, title='Number of Cells'),
+                                  xaxis=dict(tickwidth=1, title='Transformed X'),
-                         yaxis=dict(tickwidth=1, title='Number of Layers'),
+                                  yaxis=dict(tickwidth=1, title='transformed Y'),
-                         zaxis=dict(tickwidth=1, title='Position -pX')),
+                                  zaxis=dict(tickwidth=1, title='Epoch')),
                        title='{} - Latent Trajectory Movement'.format('Penis'),
                        width=800, height=800,
                        margin=dict(l=0, r=0, b=0, t=0))
@@ -195,4 +190,8 @@ if __name__ == '__main__':
     in_file = args.in_file[0]
     out_file = args.out_file
 
+    with open(in_file, 'rb') as in_f:
+        experiment = dill.load(in_f)
+        plot_latent_trajectories_3D(experiment.data_storage)
+
     print('aha')