PyTorch Mobile mentions that is available for Linux platform.
Is there available some workflow to build it from source for Linux? Further is there some Linux-based examples for pre-trained model deployment and if possible, some model optimization steps for effective infer?