Created on 06-21-2015 12:26 AM - edited 09-16-2022 02:31 AM
Hi,
I am having issue with oryx training and testing data output. It seems that during map reduce step, it splits the data into training and testing, and computes X and Y based on training data. But in the output folder, the train dataset in the train folder is always the same as the original data, and the test folder is always missing any data, no matter what test-set-fraction is set. Would you mind advising me of what might be the issue?
Thanks!
Ying
Created 06-23-2015 09:16 AM
Hi Sean,
You are right. It has to do with config. I have figured it out. Thanks so much!
Ying
Created 06-21-2015 03:40 AM
Tell me more about your setup -- what's your config, how much data are you sending? do you see any log messages about "Mean average precision:" ?
Created 06-23-2015 09:16 AM
Hi Sean,
You are right. It has to do with config. I have figured it out. Thanks so much!
Ying