VictorLlu

Apologies for the delay due to the ECCV submission deadline. In compliance with our company's confidentiality policies, the original code cannot be published. This version of the code has been...

The checkpoint is the pretraining checkpoint, Please refine to issue https://github.com/fudan-zvg/RoadNet/issues/2#issuecomment-2004486882

I've made a minor modification to the image loading process: ```python img_bytes = [ get(name, backend_args=self.backend_args) for name in filename ] img = [ mmcv.imfrombytes(img_byte, flag=self.color_type) for img_byte in img_bytes...

I find it highly related to the `num_workers` > @VictorLlu FYI, I print time for data preprocessing (time1), forward propagation (time2) and back propagation (time3) in `mmengine/model/wrappers/distributed.py` when training with...

> @VictorLlu FYI, I print time for data preprocessing (time1), forward propagation (time2) and back propagation (time3) in `mmengine/model/wrappers/distributed.py` when training with 8 GPUs. It seems that the major cause...

You can just use the following block to replace pyramid block strcuture ``` self.block = nn.ModuleList([SoftmaxFreeTransformerBlock( dim=embed_dim, num_heads=num_head, drop_path=dpr[i], ratio=sr_ratios[0], conv_size=9, max_iter=newton_max_iter, kernel_method=kernel_method) for i in range(depths)]) ``` and use...

You may first install SOFT with ``` python -m pip install -e SOFT ``` Or, we have provided a pytorch version implementation, you can just **delete** C-related sentences in Python...

You would better first install SOFT with ``` python -m pip install -e SOFT ``` This will solve most of the issues caused by library `SOFT`. And then, if you...