-
Hi all, i found a very strange behaviour maybe someone can explain. Currently trying to teach face recognition and have list of Celebrities in my training data. Net did crash during learning phase wit…
-
AWS re:invent 2018 機械学習関連セッションのプレゼン資料・動画一覧 #reinvent
gistで主に作業してる。
https://gist.github.com/mayosuke/b4402e2fdc48828b3d282d672599cefc
-
> AI - 人工智能;AR - 增强现实;CV - 机器视觉;DL - 深度学习;DM - 数据挖掘;DS - 数据科学;DV - 数据可视化;IOT - 物联网;ML - 机器学习;NLP - 自然语言处理
-
Hi Adam,
Thanks for your great work! I have a quick question on the relationship with the CMU OpenFace. According to your Medium post 2 years ago [1], the original face_recognition is designed to s…
-
-
These 328 name tokens were derived from the top 100 lists in Google Trends via [this repository](https://github.com/az0/google-trend-names). The geography was set to US, and it spanned dates from 2004…
-
I have seen that the current top algorithm in the MegaFace challenge is iBug_DeepInsight, with an accuracy that corresponds with your latest update: *2018.02.13: We achieved state-of-the-art performan…
d4nst updated
6 years ago
-
Why the `forEach` closures instead of `for/in` loops? For loops are a bit more "standard" and idiomatic, so just curious here.
https://github.com/DevWithTheHair/Conference-Talks/blob/3e5acb4ea6b1e1…
-
What's the type of `allImages`? The failure closure had the types explicitly listed; might be nice to do the same so it's easy to look up the response type via google.
https://github.com/DevWithThe…
-
Probably a quirk of the API, but if the response variable is `allImages` with `images` with `faces` — how can one input image URL spawn many images?
In actual use will `allImages` always only have …