diff --git a/example/zillow_regression_sparse/README.MD b/example/zillow_regression_sparse/README.MD
index cf9a87f..00bd784 100644
--- a/example/zillow_regression_sparse/README.MD
+++ b/example/zillow_regression_sparse/README.MD
@@ -28,7 +28,7 @@ The architecture of this ensemble is illustrated bellow.
 12. From now on, everything is optional. You can get a score of 0.0643+ (top 3% at the time) if you blend with the output of [another script](https://www.kaggle.com/davidfumo/boosted-trees-lb-0-0643707/output).
 13. Download **sub20170821_221910.csv** from the link in (12) and place it into the base directory.
 14. You can generate a submission combining the 2 submissions via executing the **combine_subs.py** as `python combine_subs.py`. It will be saved to *output_dataset2_merged.csv*
-15. Again optionally you can get the great genetic programming script from [here](https://www.kaggle.com/scirpus/genetic-programming-lb-0-0643904/output) and unzip it to get to 0.06425+(top 40 the time of post) . The file tha comes out after unzipping is teh **xxx.csv**.
+15. Again optionally you can get the great genetic programming script from [here](https://www.kaggle.com/scirpus/genetic-programming-lb-0-0643904/output) and unzip it to get to 0.06425+(top 40 the time of post) after merging with the previous submission. The file tha comes out after unzipping is teh **xxx.csv**.
 16.  You can generate a submission combining the 2 submissions via executing the **combine_subs_v2.py** as `python combine_subs_v2.py`. It will be saved to *output_dataset2_merged_v2.csv*
 
 This is how your base directory needs to look like after everything is done: