-
Hi!
Recently I bulid learnable adjacency matrices for modeling in which case connections between different sites vary at different time steps. (Each time step corresponds to a specific graph structur…
-
## 🐛 Bug
## To Reproduce
```
a=torch.randn(2,3).to_sparse().requires_grad_(True)
c=torch.randn(2,3)
sum2=c+a
loss=sum2.sum()
loss.backward()
```
results in
```
Function AddBackward0 re…
-
Hello, JiaxiongQ!
When I run test.py, I am a little bit confused about the save of dense depth image, can u tell me what u consider about the save process. ( I always like to use cv2.imsave() to sav…
-
Hi taigw,
I am trying to run dense_crf3d on single modality (only T1c).
At first I used: ```dense_crf_param['BilateralModsStds'] = (3.0)``` but that didnt work. So I used ``` dense_crf_param['Bila…
-
suppose I have a depth map (near perfect) for each image with metric scale, is there a way to use them directly for dense reconstruction?
-
代码第265行,多卡数据同步之后,cross_targets计算方式有问题,应该得考虑当前local rank。
https://github.com/FlagOpen/FlagEmbedding/blob/97f57a1b92dc68d56731a1e38a2d3aad4cd67e20/FlagEmbedding/BGE_M3/modeling.py#L265
原始是:cross_tar…
-
Hello,
While using OWLVit as a base model, I get the following error:
RuntimeError: Input type (torch.FloatTensor) and weight type (torch.cuda.FloatTensor) should be the same or input should …
-
### 🐛 Describe the bug
Exporting a VIT with flexattention I got this error https://github.com/pytorch/pytorch/issues/137759#issuecomment-2470595683
As in that thread It seems that is a distinct is…
-
from transformers import BertTokenizer, BertModel
tokenizer = BertTokenizer.from_pretrained("MODEL_NAME")
model = BertModel.from_pretrained("MODEL_NAME")
我将模型下载到了本地,MiniRBT-h256-pt文件夹下有三个文件 con…
-
您好!我遇到了这个问题:
New map created with 888 points
receive a keyframe, Frame id = 56 , KF No.1
******************* Running PointCloudMapping thread wake. *******************
No valid point in pass throu…