Attempt on a Kaggle competition, Personalized Web Search Challenge, hosted by Yandex (http://www.kaggle.com/c/yandex-personalized-web-search-challenge)
12
stars
4
forks
source link
Script to evaluate, split train=24, test=3, evaluation metric #39
"cross-validation" over random subsets of users (like bootstrap?) for evaluating our models... single script to automate entire process, including evaluation metric
The first 24 days will be local train data, and the last 3 days will be local test data.
Probably use script/file_manipulation/sample_script.py, which samples 1% users from train data
"cross-validation" over random subsets of users (like bootstrap?) for evaluating our models... single script to automate entire process, including evaluation metric
The first 24 days will be local train data, and the last 3 days will be local test data.
Probably use script/file_manipulation/sample_script.py, which samples 1% users from train data