search

dwofk / fast-depth

ICRA 2019 "FastDepth: Fast Monocular Depth Estimation on Embedded Systems"
MIT License
929 stars 188 forks source link

How can it work for Raspberry pi? #24

Open oroelipas opened 4 years ago

oroelipas commented 4 years ago

Sorry, I'm new using pytorch and tvm so maybe I understand things wrong...

As I know all models in results/tvm_compiled/ are compiled for Jetson TX2.

What do I need to run a compiled model in a Raspberry pi as in tx2_run_tvm.py ?

I guess I have to Compile the model using tvm as explained in this tutorial: https://docs.tvm.ai/tutorials/frontend/from_pytorch.html

Can anyone confirm?

dwofk commented 4 years ago

Hi @Oroelipas

You were correct in that you need to compile the model anew for a different hardware platform. As the models in this repository were compiled specifically for the TX2, they will not be compatible with the Raspberry Pi without re-compilation.

ideasxiang commented 3 years ago

@Oroelipas were you able to recompile for raspberry pi?

oroelipas commented 3 years ago

Hi @ideasxiang, currently I am working in other projects and have not made progress with this. Anyway I am listening if someone publish something about it

ideasxiang commented 3 years ago

@Oroelipas thank you for your reply, I will try recompile using the link you provided. https://docs.tvm.ai/tutorials/frontend/from_pytorch.html

microboym commented 1 year ago

@ideasxiang Did you succeed to recompile the model for raspberry pi?