Visualization service that graphs the outputs of CaImAn in live time. Servaiman is optimized for CaImAn's OnACID (Online Analysis of Calcium Imaging Data in Real Time) protocol, but since it's packaged as a separate service can work with other analysis engines with modifications. Contact tom1193 if you're interested in Servaiman for other use cases.
As OnACID runs, it generates outputs that are posted to the Servaiman service that caches the most recent outputs. A visualization app, Vis, then periodically requests the cached ouputs and renders graphs from it. We use gRPC to define the RPC calls between the service and the clients, CaImAn and the vis app. Vis is under construction.
We define a method for each major data type in CaIman's output:
| Method | Caller | Description |
|---|---|---|
| PostMeta | CaImAn | Post meta data |
| PostCn | CaImAn | Post background image |
| PostAb | CaImAn | Post a matrix encoding A and b, spatial data of components and background signal respectively. |
| PostCf | CaImAn | Post a matrix encoding C and f, temporal activity of components and background signal respectively. |
| GetMeta | Vis | Get meta data |
| GetCn | Vis | Get background image |
| GetAb | Vis | Get a matrix encoding A and b |
| GetCf | Vis | Get a matrix encoding C and f |
Servaiman requires python 2.7 or 3.4+ and gRPC. Vis requires the same dependencies (pylab, bokeh) as CaImAn, so it should be run in the CaImAn environment. For CaImAn to call Servaiman, gRPC needs to be added to the caiman virtual environment.