A-SOID
A-SOID copied to clipboard
YttriLab
An active learning platform for expert-guided, data efficient discovery of behavior.
Currently the maximum likelihood cutoff for pose files that one can use is 0.95. For most solid DLC RCNNs one would like to use up until 0.99 as a cutoff...
Hi all, Partially predescribed, but I wanted to hint towards two bugs that I came across while working with ASOID extensively now: Windows 10, most current version of ASOID a)...
**Describe the bug** 12 Keypoints are in DLC pose file. 4 where excluded during learning. No errors during learning. After training, new pose files containing the original 12 keypoints where...
**Describe the bug** When I start the streamlit app in the browser the app is launched however I get the following announcement: deprecation.showPyplotGlobalUse IS DEPRECATED. The support for global pyplot...
Hi, A-SOID team this tool is a great one for our projects. Is there any way to use the workflow in jupyter? Thank you Wulin
**Describe the bug** Initial warning of feature length not matching targets after refining samples in the Create New Dataset tab. This is followed by an error in Active Learning ('input...
I noticed that the A-SOiD demo have used the CalMS21 data set that includes pose estimation and annotation from MARS. I'm currently working with MARS and BENTO to annotate my...
I've been trying to get a model working for bottom-up recording and tracking of a mouse, with intentions of being able to automatically record grooming and rearing behaviours. However, on...
**Describe the bug** I trained a pose estimation model in SLEAP, which was then uploaded into ASOiD along with an annotation file in CSV format. I was able to pass...
**Describe the bug** At the Active Learning step, whilst being able to view the performance of the classifier with ~60 iterations, upon refreshing the page an error is encountered. Additionally,...
