Overfitting refers to a model that models the training data too well.
Underfitting refers to a model that can neither model the training data nor generalize to new data.
We have p parameters and n sample.
over fitting results from trying to estimate too many parameters from too small a sample, when p>n
if we remove one feature, we will decrease the degree of overfitting .
Subscribe to:
Post Comments (Atom)
-
Previously, I wanted to install "script" on Atom to run PHP. And there was some problem, like the firewall. So I tried atom-runner... -
I tried to commit script to bitbucket using sourcetree. I first cloned from bitbucket using SSH, and I got an error, "authentication ... -
https://github.com/boto/boto3/issues/134 import boto3 import botocore client = boto3.client('s3') result = client.list_obje...
No comments:
Post a Comment