diff --git a/CHANGELOG.md b/CHANGELOG.md
index 2e259b9..975ba79 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -2,8 +2,12 @@
 
 ## Master version
 
+## ANNZ 2.0.5 (17/6/2015)
+
 - Fixed bug in `CatFormat::addWgtKNNtoTree()`, where the weight expression for the KNN trees did not include the `ANNZ_KNN_w` weight in cases of `glob.annz["addInTrainFlag"] = True`.
 
+- Modified the condition on the MLM error estimator which is propagated to PDFs in randomized regression. Previously if an error was undefined, indicating a problem, the code would stop. Now the MLM is ignored and the code continues. The change is needed as sometimes there is a valid reason for the error to be undefined.
+
 - Other minor modifications.
 
 ## ANNZ 2.0.4 (19/3/2015)
diff --git a/README.md b/README.md
index a0745eb..dd6572a 100644
--- a/README.md
+++ b/README.md
@@ -1,4 +1,4 @@
-# ANNZ 2.0.4
+# ANNZ 2.0.5
 
 ## Introduction
 ANNZ uses both regression and classification techniques for estimation of single-value photo-z (or any regression problem) solutions and PDFs. In addition it is suitable for classification problems, such as star/galaxy classification.