Yuyao Huang

RuntimeError: sparse tensors do not have strides

1

I have encountered a problem which can be reproduced by the following minimal code: ```python import torch from spconv.pytorch import SparseConv2d, SparseConvTensor, SparseReLU from spconv.pytorch.functional import sparse_add device='cuda:0' x =...

Feature Request: return iterable instead of list

1

* Parallel TQDM version: 0.2.0 ### Description I need to post-process the results returned by pqdm(). The number of tasks is very large, so it is not acceptable to store...

我未能在 t2t_vit 的代码中找到调用 External_attention 实现的地方，能否帮忙指出？ 论文表2中提到的 T2T-Transformer ，是否是对应着 T2T_module(tokens_type=...) 参数？公开版本的代码中是否可以/如何将其替换为 EA ？ 这个 issue 里提到 https://github.com/MenghaoGuo/EANet/issues/7#issuecomment-841612601 ，使我对表 2 的结果感到困惑，恳请您拨冗阐释清楚。 非常感谢！

Two tuples as dictionary keys report a false alarm of SchemaMissingKeyError

2

I am using schema 0.7.5 and python 3.11. Following is a minimal code to reproduce ```python from schema import * data_schema = Schema({ ('map_point', 'to', 'map_polygon'): {}, ('map_polygon', 'to', 'map_polygon'):...

Suspicious implementation of multi-class version _focal_loss_cost

4

https://github.com/open-mmlab/mmdetection/blob/f78af7785ada87f1ced75a2313746e4ba3149760/mmdet/models/task_modules/assigners/match_cost.py#L252C1-L269C38 When the input is not binary, why in line 262 sigmoid is used instead of softmax? In the sigmoid case, the score is not normalized among classes IMO. Further,...

A question about the design of decoder sampler.

1

Dear author, Thank you for your excellent work! I found the solution elegant and concise in general after reading your paper. However, I have a question for the decoder sampler...