Alternative Model Loading

Is there a way to load a model from bytes of a file rather than the default load from path:
Module.load(MainActivity.assetFilePath(getApplicationContext(), “deeplabv3_scripted.pt”));

Something like: Module.loadBytes(byteArray);

Hy @syed_kamal couold you elaborate a bit more on why you’d need something like this?

It was a task in my last job, for encrypting the model and decrypting it in runtime in the form of bytes without having to save it anywhere locally.

We have support for encrypted model loading in torchserve https://github.com/pytorch/serve/blob/master/docs/management_api.md#encrypted-model-serving