Anastasis Germanidis

Thanks for doing all this work, @JeanDown123. I'll try to incorporate the filtering to autosub shortly.

Can you give more details? It's hard to tell from the diff what's actually being changed (indentation issue?)

Hi @lacouture100! The plugin should work both in Edit and Play mode. Are you getting any errors when running it Play mode?

I've been facing a similar issue. I compiled `caffe_lrcn` and pointed to it in `settings.py`, and confirmed that I'm using it by printing `sys.path`, yet I can't get example 5...

@thommiano I'm using CPU mode above. I also repeated the process on an EC2 GPU instance with CUDA 7.5, ran into the same issue.

I manually compiled with make. I was under the impression that the `download.sh` scripts are for downloading the pretrained models, not caffe.

@thommiano Compiled from @anguyen8's repo

@anguyen8: Still facing the same issue with your copy... I ended up making the script work by deleting the line with the check in `embed_layer.cpp` and recompiling.

Hi @reborn225 and @masterofshapes, you need to first create an OSC client and then send a message with the address `/server/connect` to Runway's OSC server at the port displayed within...

The most lightweight way to do this is probably to retry logging in if you get a 400 error when performing any of the bot methods. I think I'll make...