search

kaz-Anova / StackNet

StackNet is a computational, scalable and analytical Meta modelling framework
MIT License
1.32k stars 344 forks source link

How to set gpu computing for some of the algorithms? #48

Open qq609299174 opened 7 years ago

qq609299174 commented 7 years ago

I have tried some of the algorithms like svm and knn but it seems too slow to run on my computer. Are there any ways to make those algorithms running faster?

kaz-Anova commented 7 years ago

Well...These algorithms are generally slow - anywhere you run them..

However, assuming you have a better implementation of the in python, you could use the python generic module that allows you to run your own algorithms (with your own backend).

You can also do your own processing in there - have a look at this thread too

qq609299174 commented 7 years ago

Okay, I will try them then. Another problem is still on knn. I set n_neighbor: 2, 4, 8,16 for each model but seems the cv results are the same. They should be different right? mae : 0.07718227001626582 mae : 0.07718227001626582 mae : 0.07718227001626582 mae : 0.07718227001626582 I forget to tell you that I used SklearnknnRegressor.

gdragone1 commented 5 years ago

Okay, I will try them then. Another problem is still on knn. I set n_neighbor: 2, 4, 8,16 for each model but seems the cv results are the same. They should be different right? mae : 0.07718227001626582 mae : 0.07718227001626582 mae : 0.07718227001626582 mae : 0.07718227001626582 I forget to tell you that I used SklearnknnRegressor.

do you run 'main_querry_v1.py' file success? I can't run it.