chenjjcccc

Hi，have you solved this problem?

> @dekrypting-karma > > I could avoid `KeyError: 'BTS_BTC'` by replacing the for-loop inside `CoinList.__init__()` (inside `coinlist.py`) with this > > ``` > for k in ticker.keys(): > if k.startswith("BTC_")...

> > segment_anything/modeling/mask_decoder.py -- line126-127 > > ```python > > src = torch.repeat_interleave(image_embeddings, tokens.shape[0], dim=0) > > src = src + dense_prompt_embeddings > > ``` > > > > >...

> same issue. > > I noticed that in [MedSAM](https://github.com/bowang-lab/MedSAM/blob/7e86549203bda3233afc50b8c8cc41c521d88c9b/segment_anything/modeling/mask_decoder.py#L126), they do the same modification as @bach05. > > compared to repeat all the embeddings, this seems to be more...

> export 的时候带上output_op参数（试了softmax）没有问题。default 是argmax，难道这个有问题？ 我用提供的推理模型测试好像是没问题的

> 你好，请问您是怎么解决这个问题的呢？我遇到了同样的问题。 Hi, how did you solve this problem? I am experiencing the same problem. 作者好像说是数据传输方出现了问题，所以我也没有解决这个问题。我觉得你可以换用自己的数据集进行训练

> 你用自己的数据跑通了吗，我自己的数据不知道该怎么去用？向你请教一下 > […](#) > ---- Replied Message ---- | From | ***@***.***> | | Date | 7/13/2023 09:29 | | To | ***@***.***> | | Cc | ***@***.***> ,...