Comments for Quantitative thoughts

Comment on Garmin data visualization by Micha

Micha — Tue, 28 May 2013 16:45:34 +0000

Dear all,

I tried to follow you steps. In R the following function call:
workout <- system(paste("python", "~/GARMIN/run.py", "sample-activity.fit"), intern = TRUE)

created an error message via python.

File "/home/micha/GARMIN/run.py", line 22
print(';',sep='',end='')
^
IndentationError: expected an indented block
Warning message:
running command 'python ~/GARMIN/run.py sample-activity' had status 1

Why did you not receive such a message?

Comment on Trading volume forecast for an illiquid stock by gb

gb — Mon, 18 Feb 2013 20:18:09 +0000

At first glace here… if volume is a standalone of vector, looks like x = NROW(volume) and resulting cumsum on x, is causing your participation based off of number of rows vs actual volume.

Perhaps I am missing something(?).

Comment on Garmin data visualization by John Doe

John Doe — Thu, 31 Jan 2013 15:24:22 +0000

Hi Dzidas,

Thanks! I think I will develop further on this and make it my preferred workout data analysis tool.

I see that I also got your name wrong. Well now you should have my email address. I don’t know what mistake I made the first time.

Great initiative!
- John

Comment on Garmin data visualization by Dzidorius Martinaitis

Dzidorius Martinaitis — Wed, 30 Jan 2013 12:17:29 +0000

Hi John,

I sent an email to you, but the address is wrong.

All these functions (rollmean, last) belongs to xts package:
install.packages(‘xts’)
require(‘xts’)

Dzidas

Comment on Garmin data visualization by John Doe

John Doe — Sat, 26 Jan 2013 00:19:35 +0000

Great job, Matt!

I’m not totally satisfied with pytrainer, and like you I’m used to R. Would you mind helping making the code more general? I downloaded your code, and I can get the data into R. But then I have a couple of doubts because you’re using functions that are not defined. Like na.locf, rollmean, last, and probably more. Would you mind uploading these as well?

Also, you use “sample….fit” in run.py. I changed this to
import sys
activity = Activity(sys.argv[1])
so that the file name is an input to the script.

Thank you!
- John

Comment on Machine learning for hackers by A Theory of Data, Clyde Coombs | Civil Statistician

A Theory of Data, Clyde Coombs | Civil Statistician — Thu, 06 Dec 2012 22:57:06 +0000

[...] these techniques seem to be related to multidimensional scaling, now used in political science for voting similarity analyses (that example is based on Machine Learning for Hackers, which I’m just about to read too) or [...]

Comment on My first competition at Kaggle by Data Scientist or DevOps or Programmer « The World's Oldest Intern

Data Scientist or DevOps or Programmer « The World's Oldest Intern — Sun, 18 Nov 2012 09:31:51 +0000

[...] http://www.investuotojas.eu/2012/07/02/my-first-competition-at-kaggle/ [...]

Comment on Machine learning for hackers by Vita

Vita — Thu, 25 Oct 2012 12:48:23 +0000

Thanks for the tip, this book looks exactly like the ML crash course I need!

Voting results can perhaps be obtained more easily. I once sent a polite inquiry to the admins of my country’s parliament servers (I come from the Czech Republic) and they promptly sent back fresh database dumps in CSV

Comment on Machine learning for hackers by Sirvydas

Sirvydas — Wed, 24 Oct 2012 06:13:03 +0000

An interesting note about the outliers. This could be an addition to the http://manoseimas.lt project.

Comment on ai-class.com vs ml-class.com by Dzidorius Martinaitis

Dzidorius Martinaitis — Fri, 12 Oct 2012 10:25:59 +0000

Kris,
you are mixing past with present