Hongwei Niu

> This is a classic (accidental) typographic attack. > >  > > See the _Fallacies of abstraction_ section of [OpenAI - Multimodal neurons in artificial neural networks](https://openai.com/research/multimodal-neurons)....

oh, thanks.

> You can use `predict_torch` to give multiple bounding boxes as input prompts: > > ```python > input_boxes = torch.tensor([box_1, box_2], device=mask_predictor.device) > transformed_boxes = mask_predictor.transform.apply_boxes_torch(input_boxes, image.shape[:2]) > masks, iou_predictions,...

> So this is my Solution: > > ``` > input_points = [] > > input_boxes = [] > input_label = [] > > for groupedPoints in groupedPointsByBoxes.items(): > pointsByBoxForSegmentation...

I have the same problem.

> y是为了减少节点个数，从而减少计算邻接矩阵时的计算量。 谢谢回答！

> As far as I understand: > > * Pair: number of samples (videos) > * L: length of video in frames > * T: time axis? It's always 1...

I have the same question.

> It uses the Gumble-softmax to make propagation come true. May I ask on which line of the code is it

> In short: They are used in the indicator matrix of `sigmoid_weighted_focal_loss()` in `models/segmentation.py` to not penalize predictions without a pseudo label from the filter. > > See in engine.py...