Frederik Haa

Can you tell how you changed the code to use "mobilenet_v2.mobilenet_v2_140"?

We have used the default segmentation model on a data set consisting of 160 training scenes and 40 validation scenes with four classes: grass, trees, humans and clutter. We reached...

My scenes contains an imbalance of grass meaning that many blocks only contains grass points. We've had succes with discarding blocks containing only grass, with a probability of 80%. Since...

Getting same error after adding a random UUID to the prediction: ``` { "id": 450, "data": { "audio": "/data/upload/38/62bd9e8f-RADIO_YYY_1642203600.wav" }, "annotations": [], "predictions": [ { "id": 168, "model_version": "1655812104", "created_ago":...

It works now on python 3.10! Any reason to explicitly not support pytorch 1.13?

The microphones are attached to people who move around, so they have overlapped speech in periods, and separate in others. The microphones are synchronized.

Well, my hack works fine, so not much sense in paying for consulting there. This feature request was mostly about that the interface would be nicer if you could input...

It returns nothing, i.e. nothing is bound to that combination. I also tried with bindkey | grep "^" and looked through the list, nothing with ` showed up.

Okay, I had ^` set as a custom shortcut in gnome settings. After removing that it works.

Any word on when we can expect this new whisper package?