Add dashboards for generating meshes and mesh metrics (#274)

* Implement dashboards using streamlit / bokeh

Author: stefsmeets

dash/bokeh_plots.py

@@ -0,0 +1,183 @@
+import itertools
+
+import numpy as np
+from bokeh.models import LinearColorMapper
+from bokeh.palettes import Category10_10 as catpalette
+from bokeh.palettes import Viridis256 as palette
+from bokeh.plotting import figure
+from bokeh.transform import factor_cmap
+
+from nanomesh import Image, metrics
+
+
+def image_plot(data):
+    if isinstance(data, Image):
+        data = data.image
+
+    fig = figure(tooltips=[('x', '$x'), ('y', '$y'), ('value', '@image')],
+                 match_aspect=True,
+                 border_fill_alpha=0.0,
+                 height=300)
+
+    fig.x_range.range_padding = fig.y_range.range_padding = 0
+
+    dw, dh = data.shape
+
+    # must give a vector of image data for image parameter
+    fig.image(image=[data],
+              x=0,
+              y=0,
+              dw=dw,
+              dh=dh,
+              palette='Viridis11',
+              level='image')
+    fig.grid.grid_line_width = 0.5
+    return fig
+
+
+def contour_mesh(mesher):
+    fig = figure(title='Line Mesh',
+                 tooltips=[('x', '$x'), ('y', '$y'), ('value', '@image')],
+                 match_aspect=True,
+                 border_fill_alpha=0.0)
+
+    c = mesher.contour
+
+    vert_x, vert_y = c.points.T
+
+    colors = itertools.cycle(catpalette)
+
+    for i in np.unique(c.cell_data['segment_markers']):
+        color = next(colors)
+        mask = c.cell_data['segment_markers'] != i
+
+        if mask is not None:
+            cells = c.cells[~mask.squeeze()]
+
+        lines_x = (vert_x[cells] + 0.5).tolist()
+        lines_y = (vert_y[cells] + 0.5).tolist()
+
+        fig.multi_line(lines_y,
+                       lines_x,
+                       legend_label=f'{i}',
+                       line_width=5,
+                       color=color)
+
+    fig.hover.point_policy = 'follow_mouse'
+
+    # image
+    d = mesher.image
+
+    fig.x_range.range_padding = fig.y_range.range_padding = 0
+
+    dw, dh = d.shape
+    # must give a vector of image data for image parameter
+    fig.image(image=[d],
+              x=0,
+              y=0,
+              dw=dw,
+              dh=dh,
+              palette='Viridis11',
+              level='image')
+    fig.grid.grid_line_width = 0.0
+    return fig
+
+
+def get_meshplot(mesh):
+    fig = figure(title='Meshplot',
+                 tooltips=[('Physical', '@physical'), ('(x, y)', '($x, $y)'),
+                           ('index', '$index')],
+                 x_axis_label='x',
+                 y_axis_label='y',
+                 match_aspect=True)
+
+    color_mapper = LinearColorMapper(palette=palette)
+
+    xs = mesh.points[mesh.cells][:, :, 0]
+    ys = mesh.points[mesh.cells][:, :, 1]
+    cs = mesh.cell_data['physical']
+
+    data = {
+        'x': xs.tolist(),
+        'y': ys.tolist(),
+        'physical':
+        tuple(mesh.number_to_field.get(val, str(val)) for val in cs)
+    }
+
+    factors = tuple(reversed(mesh.fields))
+
+    fig.patches(
+        'x',
+        'y',
+        source=data,
+        # line_color='white',
+        fill_color=factor_cmap('physical', palette=catpalette,
+                               factors=factors),
+        fill_alpha=0.7,
+        line_width=1)
+    fig.hover.point_policy = 'follow_mouse'
+
+    return fig
+
+
+def get_metric_hist(mesh, metric):
+    metric_vals = getattr(metrics, metric)(mesh)
+
+    hist, edges = np.histogram(metric_vals, bins=50)
+
+    fig = figure(
+        title=f'Histogram of triangle {metric}',
+        x_axis_label=f'Triangle {metric}',
+        y_axis_label='Frequency',
+        tooltips=[
+            ('value', '@top'),
+        ],
+    )
+    fig.quad(top=hist,
+             bottom=0,
+             left=edges[:-1],
+             right=edges[1:],
+             fill_color='navy',
+             line_color='white',
+             alpha=0.5)
+    fig.hover.point_policy = 'follow_mouse'
+
+    return fig
+
+
+def get_metric_2dplot(mesh, metric):
+    metric_vals = getattr(metrics, metric)(mesh)
+
+    color_mapper = LinearColorMapper(palette=palette)
+
+    fig = figure(title=f'Triplot of triangle {metric}',
+                 tooltips=[('Physical', '@physical'),
+                           (metric.capitalize(), f'@{metric}'),
+                           ('(x, y)', '($x, $y)')],
+                 x_axis_label='x',
+                 y_axis_label='y',
+                 match_aspect=True)
+
+    xs = mesh.points[mesh.cells][:, :, 0]
+    ys = mesh.points[mesh.cells][:, :, 1]
+
+    cs = mesh.cell_data['physical']
+
+    data = {
+        'x': xs.tolist(),
+        'y': ys.tolist(),
+        'label': cs.tolist(),
+        metric: metric_vals.tolist(),
+    }
+
+    fig.patches('x',
+                'y',
+                source=data,
+                fill_color={
+                    'field': metric,
+                    'transform': color_mapper
+                },
+                line_width=1)
+    fig.hover.point_policy = 'follow_mouse'
+
+    return fig