home/algorithm.md

@@ -21,12 +21,16 @@ Here the architecture is provided.
 
 ### For HOFT
 - To find the target glitch time series to be subtracted 
+- bandpass the strain time series with (10, 50) Hz 
+- 
 - use a dictionary learning 
-- train the dictionary using the whitened bandpassed (10, 50) Hz 
+- train the dictionary using the whitened band-passed (10, 50) Hz 
 - Actually, train multiple dictionaries using different glitch subsets
-- 
 
 
+
+#### Comparison between the full or band passed time series to be decomposed with the trained dictionaries 
+
 The plot below is created with the trained dictionaries using the band passed. The reconstructed time series is from the bank passed series. 
 
 ![Screen_Shot_2020-11-30_at_10.37.41_PM](uploads/a5eca19c11956b2b24f462e661d9be30/Screen_Shot_2020-11-30_at_10.37.41_PM.png)
@@ -35,10 +39,17 @@ The plot below is created with the trained dictionaries using the band passed. T
 
 ![Screen_Shot_2020-11-30_at_11.00.28_PM](uploads/04066b95480a661d8f9767c4f31d29f1/Screen_Shot_2020-11-30_at_11.00.28_PM.png)
 
-The reconstructed time series is irrespective whether using the bandpassed or full time series. (Note the dictionary must be trained using the bankpassed time series.)
+The reconstructed time series is almost irrespective of whether using the band passed or full-time series. (Note the dictionary must be trained using the bank passed time series.)
 
 ![Screen_Shot_2020-11-30_at_10.56.12_PM](uploads/6bf89813b60dbdcd66588332e683049e/Screen_Shot_2020-11-30_at_10.56.12_PM.png)
 
-##
+![Screen_Shot_2020-11-30_at_11.16.13_PM](uploads/241075bd789d70313278463d669ec4cd/Screen_Shot_2020-11-30_at_11.16.13_PM.png)
+
+![Screen_Shot_2020-11-30_at_11.14.40_PM](uploads/03582aecb0e1c8b94aee36d9936f6e05/Screen_Shot_2020-11-30_at_11.14.40_PM.png)
+
+#### Comparison of a single dictionary with a big chunk or multiple dictionaries with small chunks [link](comprehensive_comparison_stacking)
+
+
+