couple of typos

Author: qgeissmann

report/discussion.tex

@@ -178,7 +178,7 @@ \section*{Conclusion}
 The overall achieved accuracy was as high as 92\%, and although some significant structural differences were induced by prediction,
 the classifier was overall satisfying.
 In addition, the presented classifier can generate confidence values that can be used to moderate each prediction, and ultimately decide whether to trust them.
-Before considering implementation of this promising classifier is a ubiquitous software tool,
+Before considering implementation of this promising classifier as a ubiquitous software tool,
 it would be necessary to generalise its results by the inclusion of different sources of data.
 
 \section*{Availability}

report/report.tex

@@ -119,15 +119,15 @@
 
 \begin{abstract}
 
-Electroencephalography (EEG) is a very widespread technique which routinely used
+Electroencephalography (EEG) is a very widespread technique which is routinely used
 as a diagnostic tool for brain dysfunctions and to investigate neurological
 phenomenons.
 Because of its relatively non-invasive nature,
 it is widely used to determine and quantify sleep stages in humans and model
 mammals such as rodents.
 Historically, labelling of EEG recordings has been performed visually by trained
 human experts. This task is extremely tedious an quite subjective.   
-Despite recent efforts to develop a automatic classifier of sleep stages
+Despite recent efforts to develop a automatic classifier of sleep stages,
 little adoption has occurred, and manual scoring is still the standard.
 
 This study presents a high accuracy classifier of sleep stage that
@@ -136,7 +136,7 @@
 in combination with random forest classifier and was demonstrated to achieve
 an overall accuracy as high as 92\%.
 In order to optimise feature extraction and pave the way for a future software
-implementation, \pr{}, a \py{} package was also developed. \pr{} is here shown
+implementation, \pr{}, a \py{} package, was also developed. \pr{} is here shown
 to outperform alternative runtime implementation by several orders of magnitudes.
 	\\
 	\\