search

Xtra-Computing / thundersvm

ThunderSVM: A Fast SVM Library on GPUs and CPUs
Apache License 2.0
1.56k stars 216 forks source link

Process finished with exit code 139 (interrupted by signal 11: SIGSEGV) #142

Closed feynman233 closed 5 years ago

feynman233 commented 5 years ago

Hi, when I use the linear kernel for a large dataset in pycharm，the code ends at "Process finished with exit code 139 (interrupted by signal 11: SIGSEGV)", but rbf kernel is just Ok, how can I fix this? thanks very much ! My gpu is titanxp and the cuda version is 9.0

jiahuanluo commented 5 years ago

Hi, @feynman233, We have tested our newest code on some large datasets and it works fine. Did you use the newest library? If not and you are using pip install, you can update your package with pip install thundersvm==0.3.2 for Linux with Cuda 9.0, the newest version, or you can build it from the source. If yes, please provide more details and a dataset for us to reproduce the problem.

Thanks

feynman233 commented 5 years ago

@jiahuanluo Thanks! By the way, where can I find the newest release version of thundersvm from pip install, because there are no update histories on the github page.

jiahuanluo commented 5 years ago

We will put it on our github page and you can find it here next.

feynman233 commented 5 years ago

@jiahuanluo thanks very much!