Segmentation fault cbow size 600 or more

[CShulby]

I get a segmentation fault with high dimensions (600 or more) using cbow. A normal word2vec runs fine for this size but wang2vec does not. I am able to run wang2vec with skip.

here is the error:

line 1: 18929 Segmentation fault ./word2vec -train final.txt -output cbow_600 -size 600 -binary 1 -type 2

and the output from my log:

Starting training using file final.txt
Vocab size: 934966
Words in train file: 1461491292
Alpha: 0.047882 Progress: 4.24% Words/thread/sec: 15.22k

[wlin12]

Hi, Have you tried setting -cap 5 or some other value? People have reported that cwindow can be subject to exploding gradients for some datasets and large embedding sizes. Setting -cap will cap the gradient to that value. Cheers, Wang Ling

…

On May 20, 2017, at 9:29 PM, Cshulby ***@***.***> wrote: I get a segmentation fault with high dimensions (600 or more) using cbow. A normal word2vec runs fine for this size but wang2vec does not. I am able to run wang2vec with skip. here is the error: line 1: 18929 Segmentation fault ./word2vec -train final.txt -output cbow_600 -size 600 -binary 1 -type 2 and the output from my log: Starting training using file final.txt Vocab size: 934966 Words in train file: 1461491292 Alpha: 0.047882 Progress: 4.24% Words/thread/sec: 15.22k — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#8>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ACKTVhN5enkU_GnFQAiP6iDta7ZXYI2Vks5r700xgaJpZM4NhZ99>.