python - how to properly use sklearn to predict the error of a fit -

i'm using sklearn fit linear regression model data. in particular, response variable stored in array y , features in matrix x.

i train linear regression model following piece of code

    sklearn.linear_model import linearregression     model = linearregression()     model.fit(x,y) 

and seems fine.

then let's have new data x_new , want predict response variable them. can done doing

    predictions = model.predict(x_new) 

my question is, error associated prediction? understanding should compute mean squared error of model:

    sklearn.metrics import mean_squared_error     model_mse = mean_squared_error(model.predict(x),y) 

and real predictions new data should random number computed gaussian distribution mean predictions , sigma^2 = model_mse. agree , know if there's faster way in sklearn?

you want validate model on training data set. suggest exploring cross-validation submodule sklearn.cross_validation.

the basic usage is:

from sklearn.cross_validation import train_test_split  x_train, x_test, y_train, y_test = train_test_split(x, y)