Improve the presentation of the sanity check and only do as at the end of the...

Improve the presentation of the sanity check and only do as at the end of the run so that sporadic messages are not displayed in individual blocks.

pcigale/analysis_modules/pdf_analysis/__init__.py

@@ -195,6 +195,7 @@ class PdfAnalysis(AnalysisModule):
         obs.save('observations')
 
         results = self._compute(conf, obs, params)
+        print("\nSanity check of the analysis results...")
         results.best.analyse_chi2()
 
         print("\nSaving the analysis results...")

pcigale/analysis_modules/pdf_analysis/workers.py

@@ -207,11 +207,6 @@ def analysis(idx, obs):
         gbl_results.best.scaling[idx] = scaling[best_idx_z]
         gbl_results.best.index[idx] = (wz.start + best_idx_z*wz.step +
                                        gbl_models.block.start)
-    else:
-        # It sometimes happens because models are older than the Universe's age
-        print(f"No suitable model found for the object {obs.id}. It may be that"
-              f" models are older than the Universe or that your chi² are very "
-              f"large.")
 
     gbl_counter.inc()
 

pcigale/managers/results.py

@@ -308,15 +308,23 @@ class BestResultsManager(object):
          objects seems to be overconstrainted.
 
         """
+        # If no best model has been found, it means none could be properly
+        # fitted. We warn the user in that case
+        bad = self.obs.table['id'][np.isnan(self.chi2)].tolist()
+        if len(bad) > 0:
+            print(f"No suitable model found for {', '.join(bad)}. It may be "
+                  f"that models are older than the universe or that your χ² are"
+                  f" very large.")
+
         obs = [self.obs.table[obs].data for obs in self.obs.tofit]
         nobs = np.count_nonzero(np.isfinite(obs), axis=0)
         chi2_red = self.chi2 / (nobs - 1)
         # If low values of reduced chi^2, it means that the data are overfitted
         # Errors might be under-estimated or not enough valid data.
-        print(f"\n{np.round((chi2_red < 1e-12).sum() / chi2_red.size, 1)}% of "
-              f"the objects have chi^2_red~0 and "
+        print(f"{np.round((chi2_red < 1e-12).sum() / chi2_red.size, 1)}% of "
+              f"the objects have χ²_red~0 and "
               f"{np.round((chi2_red < 0.5).sum() / chi2_red.size, 1)}% "
-              f"chi^2_red<0.5")
+              f"χ²_red<0.5")
 
 
 class ResultsManager(object):