Emre Sevinç
Sounds interesting. This is a nice concept but finding the optimum solution that satisfies most of the developers can prove to be a little bit difficult.
Maybe keeping things as simple as possible is the way to go. On the other hand, if you like to dream of complexity, how about metrics such as "average time...
Maybe we should start a new project and call it GitHub über-data-mining Tool (and start running crazy experiments to understand and analyze developers working via Github) ;-) (Looking at how...
Merhaba, Detaylı analiz ve uygun örnekler için çok teşekkürler @hazarek! [turkish_pattern_table](https://github.com/emres/turkish-deasciifier/blob/b385d1bb81a7eb8afe0368896f7ea65f51507c53/turkish/deasciifier.py#L18) değişkeni doğrudan ve elle oluşturulmuş bir veri seti olmadığı için, ona doğrudan müdahale uygun olmayabilir. Deniz Yüret hocanın [burada...
@hazarek, sayenizde @aysnrgenc tarafından geliştiren sistemden haberdar oldum, hemen [README.md](https://github.com/emres/turkish-deasciifier/blob/master/README.md) dosyasına da ekleyeyim [aysnrgenc/TurkishDeasciifier](https://github.com/aysnrgenc/TurkishDeasciifier)'ı. İlgili [makalede](https://github.com/aysnrgenc/TurkishDeasciifier/blob/master/diacritic-restoration-recurrent.pdf) sinir ağları ile çalışan sistem performansı **%86** olarak belirtilmiş. (TABLE IV DIACRITIC RESOLUTION RESULT...
Merhaba @takicak , Çok teşekkürler bu hataları tespit ettiğiniz için. Doğrudan müdahale etmek istemeyişimin sebeplerini [şurada açıklamıştım](https://github.com/emres/turkish-deasciifier/issues/7#issuecomment-704094448) biraz. Bir başka sebep de, tahmin edilebileceği gibi, halihazırda kapsamlı bir test "suit"i...
Hello @dpthayer , does the solution at https://stackoverflow.com/a/44625895/236007 work for you? (Assuming you've already installed the MS Windows of GraphViz binaries from https://graphviz.gitlab.io/download/ )
As @RobAltena already said, preparing those data sets aren't trivial, but at least based on what @sfleisch wrote, I think adding these instructions to README, and maybe also as some...
I'm also confused about this. Something like a formula, based on CPU cores and RAM (per node) would make understanding and using easier I believe.