H@di

Results 43 comments of H@di

It seems there are still some undone works and [unix-socket](https://github.com/aurora/rmate/tree/unix-socket) has not been merged to _master_ till now. What should be done to make it ready? 🤔 Is https://github.com/aurora/rmate/pull/63#issuecomment-356314775 done?

Using 128 dimensions and then using UMAP to reduce the result to `x` and `y`, I ended up this: ![Figure_1](https://user-images.githubusercontent.com/24562441/108633887-07cc3900-748c-11eb-942f-c18f822c628d.png) Is this a right approach? Can I make it better?

No. How can I do that? Feed the final bidimensional result of embedding to `sklearn` or sth? 🤔 **P.S.** I saw many times this approach of feeding higher dimensional embeddings...

I'll test that later this way. 🤔 Different approaches to design objective functions is an important point. I did not dig too into UMAP. Thank you!

Hi @xgfs ! Any news on this? :-"

`convert` script generates a `node2id` map to convert IDs to sequential integers which is used in BCSR. You can change the converter script to save this mapping somewhere. https://github.com/xgfs/verse/blob/51fcfe9ac1b4c5a029091d2d41e77efe82bac8ab/python/convert.py#L67 Note...

> Note that I still sort the nodes, so the order should be deterministic. Oh yes. I missed the `sort` part. 🤦🏻‍♂️

Reproduced in Ubuntu, too.

سلام ‫دست شما درد نکنه. ‫مدتیه CI خراب شده - احتمالاً‌ به دلیل آپدیت شدن بسته‌ها - و من هم نرسیدم درستش کنم. فکر کنم باید اول رو درست کنم...

متشکر ضمناً ‫یه تمپلیت ورد بسیار زشت قدیما دانشکده صنایع ساخته بود که تمپلیت رسمی فنی حساب می‌شه. یک نسخه‌ش [این‌جا](https://github.com/sinamomken/tehran-thesis/wiki/%D9%85%D8%B3%D8%AA%D9%86%D8%AF%D8%A7%D8%AA-%D9%86%DA%AF%D8%A7%D8%B1%D8%B4-%D9%BE%D8%A7%DB%8C%D8%A7%D9%86%E2%80%8C%D9%86%D8%A7%D9%85%D9%87-%D8%AF%D8%B1-%D9%BE%D8%B1%D8%AF%DB%8C%D8%B3-%D8%AF%D8%A7%D9%86%D8%B4%DA%A9%D8%AF%D9%87%E2%80%8C%D9%87%D8%A7%DB%8C-%D9%81%D9%86%DB%8C-%D8%AF%D8%A7%D9%86%D8%B4%DA%AF%D8%A7%D9%87-%D8%AA%D9%87%D8%B1%D8%A7%D9%86) هست.