made mixture_and_plot provide a unique labeling

basicrta/util.py

@@ -670,16 +670,13 @@ def mixture_and_plot(gibbs, method, **kwargs):
     leg_labels = np.array([f'{i}' for i in uniq_labels])
     predict_labels = r.predict(np.log(predict_data))
 
-    # sorts = r.precisions_.argsort()[::-1]
-    # tinds = np.array([np.where(labels == i)[0] for i in uniq_labels],
-    #                  dtype=object)
-    # pinds = np.array([np.where(predict_labels == i)[0] for i in uniq_labels],
-    #                  dtype=object)
-    # for i in uniq_labels:
-    #     labels[tinds[i]] = sorts[i]
-    #     predict_labels[pinds[i]] = sorts[i]
-    tinds = [np.where(labels == i)[0] for i in uniq_labels]
-    pinds = [np.where(predict_labels == i)[0] for i in uniq_labels]
+    sorts = r.precisions_.argsort()[::-1]
+    tinds = np.array([np.where(labels == i)[0] for i in uniq_labels],
+                     dtype=object)
+    pinds = np.array([np.where(predict_labels == i)[0] for i in uniq_labels],
+                     dtype=object)
+    tinds = tinds[sorts]
+    pinds = pinds[sorts]
 
     train_data_inds = np.array([np.where(data == col)[0][0] for col in
                                 train_data])
@@ -749,33 +746,12 @@ def mixture_and_plot(gibbs, method, **kwargs):
     ax[1, 0].set_ylim(1e-4, 1)
 
     handles, plot_labels = ax[0, 0].get_legend_handles_labels()
-    # sorts = np.argsort([int(i) for i in plot_labels])
-    # handles = np.array(handles)[sorts]
-    # plot_labels = np.array(plot_labels)[sorts]
-    [handle.set_color(cmap(get_color(int(i)))) for i, handle in
-     zip(plot_labels, handles)]
-    [handle.set_edgecolor('k') for i, handle in zip(plot_labels, handles)]
     fig.legend(handles, plot_labels, loc='lower center',
                ncols=len(plot_labels)/2, title='cluster')
     fig.suptitle(f'{method} '+' '.join(keyvalpairs), fontsize=16)
     plt.tight_layout(rect=(0, 0.05, 1, 1))
     plt.savefig(f"{gibbs.residue}/results_{method}_{kwarg_str}.png",
                 bbox_inches='tight')
     plt.show()
-    # tparams, pparams = [], []
-    # for i in uniq_labels:
-    #     tinds = np.where(labels == i)[0]
-    #     pinds = np.where(predict_labels == i)[0]
-    #     tparams.append(trates[tinds].mean())
-    #     pparams.append(prates[pinds].mean())
-    # tindex = np.where(tparams == np.min(tparams))[0]
-    # pindex = np.where(pparams == np.min(pparams))[0]
-    # clu_rates = np.concatenate([trates[labels == tindex],
-    #                             prates[predict_labels == pindex]])
-    # all_results = [(1/clu_rates).mean(), confidence_interval(1/clu_rates)]
-    # train_results = [(1/trates[labels == tindex]).mean(),
-    #                    confidence_interval(1/trates[labels == tindex])]
-    # predict_results = [(1/prates[predict_labels == pindex]).mean(),
-    #                    confidence_interval(1/prates[predict_labels == pindex])]
     return r, all_labels