lumos

operands could not be broadcast together with shapes (8,256) (4,256,256)

2

hello，When I run,：`python train.py --dataset pcontext --model deeplab --aux --backbone resnest200` I found the following error： ``` File "/dahuafs/userdata/229288/00_deeplearning/01_pytorch/segEncoding/experiments/segmentation/train.py", line 238, in validation correct, labeled, inter, union = eval_batch(self.model, image,...

When to open source

2

Hello, when will you upload the code? I'm very interested in this work

Thank you for your outstanding work.But,I find the training code does not contain any information related to iterations and negative prompts.

About training script for flow match with mixdata

1

Could you share a reference script demonstrating how to train a flow model using the mixdata processing pipeline?

Error of MaskDINODecoder

1

https://github.com/IDEA-Research/MaskDINO/blob/main/maskdino/modeling/transformer_decoder/maskdino_decoder.py#L391 ```python src_flatten = [] mask_flatten = [] spatial_shapes = [] for i in range(self.num_feature_levels): idx=self.num_feature_levels-1-i bs, c , h, w=x[idx].shape size_list.append(x[i].shape[-2:]) spatial_shapes.append(x[idx].shape[-2:]) src_flatten.append(self.input_proj[idx](x[idx]).flatten(2).transpose(1, 2)) mask_flatten.append(masks[i].flatten(1)) src_flatten = torch.cat(src_flatten, 1)...