Teodor Chiaburu

OrienHead in YoloMask for different backbones

1

Hi! While trying to train a **YOLO-Mask** with a **Swin-Tiny** backbone, I found out that the _Overhead_ class in `yolov7/modeling/meta_arch/yolomask.py` requires a 4-dimensional input shape, while the Transformer only delivers...

Replaced `len()` with `shape[0]` for several feature attribution methods (*lime*, *occlusion*, *rise* and *smoothgrad*) to fix error `len is not well-defined for symbolic tensors`

Hello! I have retrained a ResNet on CUB, but when I run the visualization script, every pixel in the images is assigned a color mask, even those not belonging to...

Thank you for the great library. I am using it in XAI project for my PhD.

Interpretation of RLE values

2

Hello! May I ask what the **RLE** values stored for brush segmentations stand for? I am familiar with this encoding method from _COCO_, but I was used to them being...

Why does NHiTS need the target variable specified in the time_varying_unknown_reals attribute?

1

I was wondering why do I need to specify the target variable twice, when building a `TimeSeriesDataset` for a **NHiTS**? Once for the attribute 'target' and once for 'time_varying_unknown_reals'. If...

Bugfix: handle case where 'losses' is a list of tensors

1

### Description This PR fixes an issue I came across when applying NHiTS and TFT on my use case, namely that `losses` can be tensors within a list. In that...

Handling (custom) weather data

3

Hello! Thank you for your great contribution! We are trying to use Teaser for public buildings in Berlin. I have some questions regarding the integration of weather data: 1. Where...