search

margulies / topography

MIT License
0 stars 4 forks source link

Is distance matrix already normalized? #4

Open margulies opened 9 years ago

margulies commented 9 years ago

if so, do we have the scaling factor for each individual?

soligschlager commented 9 years ago

the distance matrix is normalized within the script. It's scaled to a range between 0 and 1 using sklearn.preprocessing.MinMaxScaler. The scaling factor is stdev * (max - min ) + min. Should this be saved for each individual?

margulies commented 9 years ago

It's possible to use un-normed dist, I think that would be more approriate for the case. Just comment out the line where its normed.

On Jun 21, 2015, at 11:59 PM, Sabine Oligschläger notifications@github.com wrote:

the distance matrix is normalized within the script. It's scaled to a range between 0 and 1 using sklearn.preprocessing.MinMaxScaler. The scaling factor is stdev * (max - min ) + min. Should this be saved for each individual?

— Reply to this email directly or view it on GitHub.