Mar 24 2016
There is very little distraction here in Račice, so I had some time to play with a python based data analysis tool that Greg Smith (from Quantified Rowing) developed for his Painsled CSV files.
I have slightly rewritten it to better match my workflow (sorry, Greg), but more importantly I have added a TCX Parser that parses my CrewNerd files. It took me some time to get up to speed with parsing XML. First time I worked with the lxml package and I needed to learn the syntax of “xpath”, which was greatly complicated because my network connection broke down on Wednesday morning.
Anyway, I succeeded in writing something that converts a TCX file to a “painsled style” CSV file and here is Tuesday morning’s workout in a first crude graph. I need to smoothen the pace graph (probably by a running average) a bit to get the noise level down. Also, I need to calculate useful stuff like meters per stroke etc. The colorful bar graph produces some strange red spikes during the interval part. Not sure what that is.
But for that I am on common ground. No problems to work with the data once I have them.
The moving average thing took me 5 minutes. Updated graph: