clean scripts

@leo.bouscarrat @luc.giffon

Il reste encore à ajouter plus de plots dans compute_results.py (pour l'instant il y a un exemple qui plot les loss sur train/dev/test).

Question : Dans le notebook il y a une version de la prédiction calculée avec une division de la somme des poids, et une version sans. C'était laquelle la bonne déjà ?

EDIT 1 - Exemple d'utilisation :

python code/train.py --dataset_name=california_housing --normalize_D --dataset_normalizer=standard --extracted_forest_size 10 20 30 --random_seed_number=2
python code/compute_results.py --experiment_ids 1

code/bolsonaro/visualization/plotter.py

@@ -6,17 +6,14 @@ from sklearn.neighbors.kde import KernelDensity
 class Plotter(object):
 
     @staticmethod
-    def weight_density(weights):
-        """
-        TODO: to complete
-        """
+    def weight_density(weights, X, file_path):
         X_plot = [np.exp(elem) for elem in weights]
         fig, ax = plt.subplots()
 
         for kernel in ['gaussian', 'tophat', 'epanechnikov']:
             kde = KernelDensity(kernel=kernel, bandwidth=0.5).fit(X_plot)
             log_dens = kde.score_samples(X_plot)
-            ax.plot(X_plot[:, 0], np.exp(log_dens), '-',
+            ax.plot(X_plot, np.exp(log_dens), '-',
                     label="kernel = '{0}'".format(kernel))
 
         ax.legend(loc='upper left')
@@ -24,4 +21,46 @@ class Plotter(object):
 
         ax.set_xlim(-4, 9)
         ax.set_ylim(-0.02, 0.4)
-        plt.show()
+        fig.savefig(file_path, dpi=fig.dpi)
+        plt.close(fig)
+
+    @staticmethod
+    def plot_mean_and_CI(ax, mean, lb, ub, x_value, color_mean=None, facecolor=None, label=None):
+        # plot the shaded range of the confidence intervals
+        ax.fill_between(x_value, ub, lb, facecolor=facecolor, alpha=.5)
+        # plot the mean on top
+        ax.plot(x_value, mean, c=color_mean, label=label)
+
+    @staticmethod
+    def plot_losses(file_path, all_experiment_scores, x_value, xlabel, ylabel, all_labels, title):
+        fig, ax = plt.subplots()
+
+        n = len(all_experiment_scores)
+
+        colors = Plotter.get_colors_from_cmap(n)
+
+        for i in range(n):
+            experiment_scores = list(all_experiment_scores[i].values())
+            mean_experiment_scores = np.average(experiment_scores, axis=0)
+            std_experiment_scores = np.std(experiment_scores, axis=0)
+            Plotter.plot_mean_and_CI(
+                ax=ax,
+                mean=mean_experiment_scores,
+                lb=mean_experiment_scores + std_experiment_scores,
+                ub=mean_experiment_scores - std_experiment_scores,
+                x_value=x_value,
+                color_mean=colors[i],
+                facecolor=colors[i],
+                label=all_labels[i]
+            )
+
+        plt.xlabel(xlabel)
+        plt.ylabel(ylabel)
+        plt.title(title)
+        plt.legend(loc='upper right')
+        fig.savefig(file_path, dpi=fig.dpi)
+        plt.close(fig)
+
+    @staticmethod
+    def get_colors_from_cmap(n_colors, colormap_name='nipy_spectral'):
+        return [plt.get_cmap(colormap_name)(1. * i/n_colors) for i in range(n_colors)]