Truncate when converting to images; don't show error/targets by defau…

…lt; Network.get_weights(name); all widget pages are same length; default colormap for all

conx/layers.py

@@ -74,7 +74,7 @@ def __init__(self, name, *args, **params):
         self.vshape = None
         self.image_maxdim = None
         self.visible = True
-        self.colormap = None
+        self.colormap = "RdGy"
         self.minmax = None
         self.model = None
         self.decode_model = None
@@ -223,7 +223,7 @@ def make_image(self, vector, config={}):
         minmax = config.get("minmax")
         if minmax is None:
             minmax = self.get_minmax(vector)
-        vector = self.scale_output_for_image(vector, minmax)
+        vector = self.scale_output_for_image(vector, minmax, truncate=True)
         if len(vector.shape) == 1:
             vector = vector.reshape((1, vector.shape[0]))
         size = config["pixels_per_unit"]
@@ -242,12 +242,12 @@ def make_image(self, vector, config={}):
         image = image.resize((new_height, new_width))
         return image
 
-    def scale_output_for_image(self, vector, minmax):
+    def scale_output_for_image(self, vector, minmax, truncate=False):
         """
         Given an activation name (or something else) and an output
         vector, scale the vector.
         """
-        return rescale_numpy_array(vector, minmax, (0,255), 'uint8')
+        return rescale_numpy_array(vector, minmax, (0,255), 'uint8', truncate=truncate)
 
     def make_dummy_vector(self):
         """

conx/network.py

@@ -135,10 +135,10 @@ def __init__(self, name: str, *sizes: int, **config: Any):
             "arrow_width": "2",
             "border_width": "2",
             "border_color": "blue",
-            "show_targets": True,
+            "show_targets": False,
+            "show_errors": False,
             "minmax": None,
             "colormap": None,
-            "show_errors": True,
             "pixels_per_unit": 1,
             "pp_max_length": 20,
             "pp_precision": 1,
@@ -1002,6 +1002,14 @@ def get_test_target(self, i):
                 targets.append(list(self.test_targets[c][i]))
             return targets
 
+    def get_weights(self, layer_name):
+        """
+        Get the weights from the model in an easy to read format.
+        """
+        weights = [layer.get_weights() for layer in self.model.layers
+                   if layer_name == layer.name][0]
+        return [m.tolist() for m in weights]
+
     def propagate(self, input, batch_size=32):
         """
         Propagate an input (in human API) through the network.
@@ -1777,7 +1785,7 @@ def get_train_targets_length(self):
         else:
             return self.train_targets.shape[0]
 
-    def dashboard(self, width="100%", max_height="550px", iwidth="800px"): ## FIXME: iwidth hack
+    def dashboard(self, width="100%", height="550px", iwidth="960px"): ## FIXME: iwidth hack
         """
         Build the dashboard for Jupyter widgets. Requires running
         in a notebook/jupyterlab.
@@ -1855,7 +1863,7 @@ def prop_one(button):
             update_slider_control({"name": "value"})
 
         net_svg = HTML(value=self.build_svg(), layout=Layout(
-            width=width, height="100%", max_height=max_height, overflow_x='auto',
+            width=width, height=height, overflow_x='auto',
             justify_content="center"))
         button_begin = Button(icon="fast-backward", layout=Layout(width='100%'))
         button_prev = Button(icon="backward", layout=Layout(width='100%'))
@@ -1894,12 +1902,12 @@ def prop_one(button):
 
         # Put them together:
         control = VBox([control_select, control_slider, control_buttons], layout=Layout(width='100%'))
-        net_page = VBox([net_svg, control], layout=Layout(width='100%'))
-        graph_page = VBox()
-        analysis_page = VBox()
-        camera_page = VBox([Button(description="Turn on webcamera")])
-        help_page = HTML('<iframe style="width: %s" src="https://conx.readthedocs.io" width="100%%" height="%s"></frame>' % (iwidth, max_height),
-                         layout=Layout(width="100%"))
+        net_page = VBox([net_svg, control], layout=Layout(width='100%', height=height))
+        graph_page = VBox(layout=Layout(width='100%', height=height))
+        analysis_page = VBox(layout=Layout(width='100%', height=height))
+        camera_page = VBox([Button(description="Turn on webcamera")], layout=Layout(width='100%', height=height))
+        help_page = HTML('<iframe style="width: %s" src="https://conx.readthedocs.io" width="100%%" height="%s"></frame>' % (iwidth, height),
+                         layout=Layout(width="100%", height=height))
         net_page.on_displayed(lambda widget: update_slider_control({"name": "value"}))
         tabs = [("Network", net_page), ("Graphs", graph_page), ("Analysis", analysis_page),
                 ("Camera", camera_page), ("Help", help_page)]

conx/utils.py

@@ -18,6 +18,7 @@
 # Boston, MA 02110-1301  USA
 
 import numbers
+import numpy as np
 from keras.utils import to_categorical
 
 #------------------------------------------------------------------------
@@ -84,15 +85,18 @@ def valid_vshape(x):
     # vshape must be a single int or a 2-dimensional tuple
     return valid_shape(x) and (isinstance(x, numbers.Integral) or len(x) == 2)
 
-def rescale_numpy_array(a, old_range, new_range, new_dtype):
+def rescale_numpy_array(a, old_range, new_range, new_dtype, truncate=False):
     """
     Given a vector, old min/max, a new min/max and a numpy type,
     create a new vector scaling the old values.
     """
     assert isinstance(old_range, (tuple, list)) and isinstance(new_range, (tuple, list))
     old_min, old_max = old_range
     if a.min() < old_min or a.max() > old_max:
-        raise Exception('array values are outside range %s' % (old_range,))
+        if truncate:
+            a = np.clip(a, old_min, old_max)
+        else:
+            raise Exception('array values are outside range %s' % (old_range,))
     new_min, new_max = new_range
     old_delta = old_max - old_min
     new_delta = new_max - new_min