The OCR approach is rephrased as Segmentation Transformer: https://arxiv.org/abs/1909.11065. This is an official implementation of semantic segmentation for HRNet. https://arxiv.org/abs/1908.07919
Dear author,
when I try to train the model in RTX3090, an issue is that it could just be worked when I use 2 GPUs. If I use 4 GPUs, The server will automatically restart after training a few iterations. I wonder if you know some suggestions.
Dear author, when I try to train the model in RTX3090, an issue is that it could just be worked when I use 2 GPUs. If I use 4 GPUs, The server will automatically restart after training a few iterations. I wonder if you know some suggestions.