Cut out extra text that has been edited out of videos - mostly contextual information about the datasets and explanatory text interpreting the plots.