Reservoir states visualisation

echotorch/utils/__init__.py

@@ -4,10 +4,10 @@
 # Imports
 from .error_measures import nrmse, nmse, rmse, mse, perplexity, cumperplexity
 from .utility_functions import spectral_radius, deep_spectral_radius, normalize, average_prob, max_average_through_time
-from .visualisation import show_3d_timeseries, show_2d_timeseries, show_1d_timeseries
+from .visualisation import show_3d_timeseries, show_2d_timeseries, show_1d_timeseries, neurons_activities_1d, neurons_activities_2d, neurons_activities_3d
 
 __all__ = [
     'nrmse', 'nmse', 'rmse', 'mse', 'perplexity', 'cumperplexity', 'spectral_radius', 'deep_spectral_radius',
     'normalize', 'average_prob', 'max_average_through_time', 'show_3d_timeseries', 'show_2d_timeseries',
-    'show_1d_timeseries'
+    'show_1d_timeseries', 'neurons_activities_1d', 'neurons_activities_2d', 'neurons_activities_3d'
 ]

echotorch/utils/visualisation.py

@@ -8,6 +8,103 @@
 from mpl_toolkits.mplot3d import Axes3D
 
 
+# Display neurons activities on a 3D plot
+def neurons_activities_3d(stats, neurons, title, timesteps=-1, start=0):
+    """
+    Display neurons activities on a 3D plot
+    :param stats:
+    :param neurons:
+    :param title:
+    :param timesteps:
+    :param start:
+    :return:
+    """
+    # Fig
+    ax = plt.axes(projection='3d')
+
+    # Two by two
+    n_neurons = neurons.shape[0]
+    stats = stats[:, neurons].view(-1, n_neurons // 3, 3)
+
+    # Plot
+    if timesteps == -1:
+        time_length = stats.shape[0]
+        ax.plot3D(stats[:, :, 0].view(time_length).numpy(), stats[:, :, 1].view(time_length).numpy(), stats[:, :, 2].view(time_length).numpy(), 'o')
+    else:
+        ax.plot3D(stats[start:start + timesteps, :, 0].numpy(), stats[start:start + timesteps, :, 1].numpy(), stats[start:start + timesteps, :, 2].numpy(), 'o', lw=0.5)
+    # end if
+    ax.set_xlabel("X Axis")
+    ax.set_ylabel("Y Axis")
+    ax.set_zlabel("Z Axis")
+    ax.set_title(title)
+    plt.show()
+    plt.close()
+# end neurons_activities_3d
+
+
+# Display neurons activities on a 2D plot
+def neurons_activities_2d(stats, neurons, title, colors, timesteps=-1, start=0):
+    """
+    Display neurons activities on a 2D plot
+    :param stats:
+    :param neurons:
+    :param title:
+    :param timesteps:
+    :param start:
+    :return:
+    """
+    # Fig
+    fig = plt.figure()
+    ax = fig.gca()
+
+    # Two by two
+    n_neurons = neurons.shape[0]
+
+    # For each plot
+    for i, stat in enumerate(stats):
+        # Stats
+        stat = stat[:, neurons].view(-1, n_neurons // 2, 2)
+
+        # Plot
+        if timesteps == -1:
+            ax.plot(stat[:, :, 0].numpy(), stat[:, :, 1].numpy(), colors[i])
+        else:
+            ax.plot(stat[start:start + timesteps, :, 0].numpy(), stat[start:start + timesteps, :, 1].numpy(), colors[i])
+        # end if
+    # end for
+    ax.set_xlabel("X Axis")
+    ax.set_ylabel("Y Axis")
+    ax.set_title(title)
+    plt.show()
+    plt.close()
+# end neurons_activities_2d
+
+
+# Display neurons activities
+def neurons_activities_1d(stats, neurons, title, timesteps=-1, start=0):
+    """
+    Display neurons activities
+    :param stats:
+    :param neurons:
+    :return:
+    """
+    # Fig
+    fig = plt.figure()
+    ax = fig.gca()
+
+    if timesteps == -1:
+        ax.plot(stats[:, neurons].numpy())
+    else:
+        ax.plot(stats[start:start + timesteps, neurons].numpy())
+    # end if
+
+    ax.set_xlabel("Timesteps")
+    ax.set_title(title)
+    plt.show()
+    plt.close()
+# end neurons_activities_1d
+
+
 # Show 3D time series
 def show_3d_timeseries(ts, title):
     """
@@ -26,6 +123,7 @@ def show_3d_timeseries(ts, title):
     ax.set_zlabel("Z Axis")
     ax.set_title(title)
     plt.show()
+    plt.close()
 # end show_3d_timeseries
 
 
@@ -46,11 +144,12 @@ def show_2d_timeseries(ts, title):
     ax.set_ylabel("Y Axis")
     ax.set_title(title)
     plt.show()
+    plt.close()
 # end show_2d_timeseries
 
 
 # Show 1D time series
-def show_1d_timeseries(ts, title):
+def show_1d_timeseries(ts, title, start=0, timesteps=-1):
     """
     Show 1D time series
     :param ts:
@@ -61,8 +160,13 @@ def show_1d_timeseries(ts, title):
     fig = plt.figure()
     ax = fig.gca()
 
-    ax.plot(ts[:, 0].numpy())
+    if timesteps == -1:
+        ax.plot(ts[:, 0].numpy())
+    else:
+        ax.plot(ts[start:start+timesteps, 0].numpy())
+    # end if
     ax.set_xlabel("X Axis")
     ax.set_title(title)
     plt.show()
+    plt.close()
 # end show_1d_timeseries