elephantum

7 ms·

61.
by elephantum 7y ago
Postgresql is quite good easy to integrate alternative to Elastic. It has built-in fulltext search: https://www.postgresql.org/docs/12/textsearch.html
62.
by elephantum 7y ago
> At the end of the day, I cant run ANY accelerated workloads using Intel graphics or AMD ....because there's simply no software support anywhere. You can, there’s an Intel framework: OpenVINO that is targeting Intel hardware, proce
63.
by elephantum 7y ago
I don't understand, why isn't there more excitement about this. This is huge in my opinion. Now you can get any environment in container on your local machine, with working Intellisense and all the IDE integrations.
64.
by elephantum 8y ago
YOLO is a combination of backend encoder and head. Backend (Darknet) is unique to YOLO, that would be the main difference with SSD, if I'm not mistaken.
65.
by elephantum 8y ago
Interesting idea! Though I have a question: in order to calculate C you need a way to attribute proposal and ground truth. It's trivial in case when there's only one instance of each class in the image. But how does it work, when
66.
by elephantum 8y ago
There’s a project that compiles ONNX models: https://onnc.ai
67.
by elephantum 8y ago
Like Tensor Processing Units?
68.
by elephantum 8y ago
It means you have two extra special cases: 1) list is empty; 2) left edge is the right edge (I.e. length is 1)
69.
by elephantum 8y ago
It's amusing, that you can substitute every reference from Ruby to Python and Rails to Django and article would still make sense. Except for the part about running in production. Seems like Python is a bit ahead there.
70.
by elephantum 8y ago
Is it possible that they foresee problems with Amazon valuation and this a way to save half of the assets they have? Whether it's true will depend on what his wife is about to do with the shares.
71.
by elephantum 8y ago
That's the offroad car I've been dreaming about!