Compatible with torch.export?

pytorch-labs / segment-anything-fast

A batched offline inference oriented version of segment-anything

Apache License 2.0

1.21k stars 72 forks source link

Closed dbbert closed 11 months ago

dbbert commented 1 year ago

Is this implementation compatible with torch.export? I want to deploy SAM in a serverless fashion but the compile step causes long cold start times.

msaroufim commented 1 year ago

The model has 0 graph breaks so indeed torch.export should work, if you run into any issues please let us know!

cpuhrsch commented 11 months ago

@dbbert I'm going to close this now since the issue seems to be resolved, but please reopen if that's not the case.