You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hello,
Awesome project!
It would be great to have this feature. In our project, a process loads models fetched from the network but we're currently restricted to only supporting models <2GB because storing/loading models from disk is not an option for us.
Cheers!
The text was updated successfully, but these errors were encountered:
kckeiks
changed the title
Load external data from memory
Onnx - Load external data from memory
Mar 13, 2024
Could you have a look at #1358 ? I think by replacing the data_provider on the Onnx structure by your own implementation, you would be able to redirect the data access to wherever you want. Please tell me what you think.
Could you have a look at #1358 ? I think by replacing the data_provider on the Onnx structure by your own implementation, you would be able to redirect the data access to wherever you want. Please tell me what you think.
Thanks for taking the time, it looks good! I think it will work for our case since we have the model and external data in memory so we could just implement ModelDataResolver. Looking forward to trying it!
Hello,
Awesome project!
It would be great to have this feature. In our project, a process loads models fetched from the network but we're currently restricted to only supporting models <2GB because storing/loading models from disk is not an option for us.
Cheers!
The text was updated successfully, but these errors were encountered: