only same output

[yunseongcho]

I'm trying to learn ffm model by this dataset

This dataset has approximately 10 million instance in size. and the code is below

import xlearn as xl ffm_model = xl.create_ffm() ffm_model.setTrain("./train.txt") param = {'task':'reg', 'lr':0.2, 'lambda':0, 'metric':'rmse'}

ffm_model.setTXTModel('./model.txt') ffm_model.fit(param, "./model.out") ffm_model.setTest("./test.txt") ffm_model.predict("./model.out", "./output.txt")

This is a variation of the code example in xlearn documentation.

But always outputs are same value. I tried to change parameter, and tried to change a little in dataset. but the problem isn't solved.. help me please.. and I can't speak Chinese plz comments in english..

[myeonghak]

same issue here. the output varies only 0.01 or something..

[myeonghak]

혹시 아직 해결 못하셨다면 제가 최근에 올린 이슈를 참고해 주시기 바랍니다. 제 경우에는 전처리 후 라벨 자리에 있어야할 데이터가 잘못될 경우에 저런 증상이 나오더라구요

[ghost]

Have you solve the problem? I meet the same situation

[myeonghak]

Have you solve the problem? I meet the same situation

please refer to this issue I posted:

https://github.com/aksnzhy/xlearn/issues/351

hope this may help!