Mercurial > hg > smacpy
changeset 19:08f6c2948349
update readme now crossvalidating
| author | Dan Stowell <danstowell@users.sourceforge.net> |
|---|---|
| date | Fri, 07 Dec 2012 12:13:56 +0000 |
| parents | 6a34565c8a74 |
| children | 7d6a647a2d44 |
| files | README.md smacpy.py |
| diffstat | 2 files changed, 2 insertions(+), 3 deletions(-) [+] |
line wrap: on
line diff
--- a/README.md Fri Dec 07 12:02:00 2012 +0000 +++ b/README.md Fri Dec 07 12:13:56 2012 +0000 @@ -30,9 +30,7 @@ If you invoke the script from the commandline (e.g. "python smacpy.py") it will assume there is a folder called "wavs" and inside that folder are multiple WAV files, each of which has an underscore in the filename, and the class label is the text BEFORE the underscore. -It will train a model using the wavs, and then test it on the same wavs. -This can be a useful test that the script is working; -however, remember that for proper evaluation you must test using files independent from the training files. +It will train a model using the wavs, and then test it on the same wavs (dividing the collection up so it can do a "crossvalidated" test). To train and test on different folders, you can run it like this:
--- a/smacpy.py Fri Dec 07 12:02:00 2012 +0000 +++ b/smacpy.py Fri Dec 07 12:13:56 2012 +0000 @@ -200,6 +200,7 @@ totcorrect, tottotal = (0,0) # Then we go through, each time training on all-but-one and testing on the one left out for index in range(numfolds): + print("Fold %i of %i" % (index+1, numfolds)) chosenfold = folds[index] alltherest = {} for whichfold, otherfold in enumerate(folds):