Closed Gh0st01 closed 6 years ago
Can you try adding this to your java command ? java -Xmx6048m -jar . This tells it to use 6 GB of RAM. You can add more if you want. It could be that due to memory the process becomes mush slower (trying to allocate memory) . See if problem persists with this,
Thank you it worked like a charm !
I don't really understand why since the process was using only 4GB ! But i used 8GB with your arg and it worked in 12sec ! I'll edit my original post to remove useless informations.
Thanks !
Hello, First of all, thanks for your awesome work.
I'm facing a strange issue.
When i train my algo with this command :
~/myfolder/StackNet $ java -jar stacknet.jar train task=classification sparse=false, s_head=false, model=/home/user/mymodelfolder/model_final pred_file=/home/user/myfolder/StackNet/apred.csv train_file=/home/user/myfolder/StackNet/trainalgo.csv test_file=/home/ghost/myfolder/StackNet/testalgo.csv test_target=true, params=/home/ghost/myfolder/StackNet/aparams.txt verbose=true threads=7 metric=auc stackdata=false, seed=2 folds=5
everything works, train and testing on my testfile but when i want to test my trained model with this command:
~/myfolder/StackNet $ java -jar stacknet.jar predict sparse=false, has_head=false, model=/home/user/mymodelfolder/model_final pred_file=/home/user/myfolder/StackNet/apred.csv test_file=/home/user/myfolder/StackNet/testalgo.csv test_target=true verbose=true metric=auc
I got this error :