On Sun, Aug 31, 2014 at 4:47 PM, sara khodeir <[email protected]> wrote:
> I'm curious what the difference between the predictors and features are in
> the logistic regression model, what is the point of having features if it
> predicts using the predictors.
>

Features are the data points of your data-set. Predictors are the data
points you choose for learning, from the features.


> Also,I am trying to include more features than can fit into the command
> line so I wrote a file of the parameters and I'm trying to run it like this
> ./mahout "filepath" but this doesn't work. Does anyone know how I can do
> this?

What version of Mahout are you using?
What exact commands are you running?

You can trail logistic regression like so:

$ mahout trainlogistic --input donut.csv --output ./model --target
color --categories 2 --predictors x y --types numeric --features 20
--passes 100 --rate 50

And then evaluate it like so:

$ mahout runlogistic --input donut-test.csv --model ./model --auc --confusion

It is not clear with your description, what steps you are following.

Are you referring to "--predictors x y" part of the command line arguments?


/tuxdna

Reply via email to