Bowen
Bowen
I have the same confusion about this problem, my 252Mb wiki small dump ends with only 18Mb, did you do a cross-check to valid the output result?
Seems the downloading url is not working again even change the http to https does not solve the problem. Any way to fix this ?@AngledLuffa
> Are you using Chrome? I've sometimes had issues where Chrome doesn't want to follow a link to something "unsafe" like a zip file. Yep, I am using Chrome.,but I...
The GPU did not improve any performance, I guess it's because the data still have to be processed by the spacy first.
If you initiate 2 nlp modules in spacy, one is normal modules and another one is a module that combined with the benepar, when you process sentences, the normal modules...
Sorry for my late reply, sincerely thank your detailed explaination! ;)
Hi, for your paper, I still have a question. The pre-calculated soft_target for duration/frequency/boundary is in pattern_extraction.py. This value is calculated as illustrated in your paper in section 3.4. However,...
> What are the output values in your version? I will try to find how I computed it exactly. Nonetheless, these are hard-coded distributions and they should not "cheat" anything....
The time dimension that is far from the target will have a small probability value, which will cause the corresponding softmax to be 1. Additionally, the target label has a...
> It seems that the standard deviations are different, where mine is much smaller so the peak value has a larger probability (0.585), and converges to 0 much faster (a...