Matt Whiteway
Matt Whiteway
Hi @faezeamin , I have not tried the multi-gpu training in several years - I can test this out on my end after the thanksgiving break and get back to...
@faezeamin sorry for not getting to this yet, haven't forgotten about it though
Hi @Zeev1988 sorry for the slow reply. No, the current codebase is not really designed for that type of task. However, the temporal convolution network (TCN) in general could be...
@ebatty did you fix this by any chance in your recent round of updates?
@hummuscience so sorry I missed this issue! Were you able to get this working? I'm pretty sure this has to do with the sequence length being 500, which is potentially...
also of note, this repo is quite out of date now, and the plan is to do a major upgrade at the beginning of the fall to match a lot...
Really depends on the species/behaviors/etc., but I would say minimum a few thousand labels per behavior class. You can also check out this paper we published recently (using the daart...
TL;DR the TCN model is a good one, but still needs lots of labeled data
@hummuscience just wanted to alert you that we're archiving this repo soon and moving over to our new Lightning Action repo, in case you're sitll interested in this: https://github.com/paninski-lab/lightning-action/
@hummuscience thanks for the kind words! We currently do not offer any pretrained models, though we hope to start providing some within the year. Once DLC releases the dataset used...