Some questions on training downstream face parsing tasks #17
Description
Has anyone used the author's pre-trained model to train a downstream task for face parsing? RuntimeError: CUDA error: device-side assert triggered when using multi-GPU training. Has anyone encountered this? Also I want to convert pth model to onnx model, has anyone done it?