Michel van Lier

Is there a YOLOX P2 model?

5

Is there an option or implementation of a YOLOX model with an extra scale output for small objects as also available in the original YoloV5: [https://github.com/ultralytics/yolov5/blob/master/models/hub/yolov5-p2.yaml](https://github.com/ultralytics/yolov5/blob/master/models/hub/yolov5-p2.yaml)

[QST] Is it possible to exchange data on GPU with OpenCV CUDA?

16

Is it possible to exchange data on GPU with OpenCV CUDA? In other words, I want to perform operations with OpenCV on the GPU with CUDA, and after that (instead...

Generator extremely slow

1

Hi, I have used your generator, for video clips of resolution: 640,480 (mp4), which are between 6sec - 5min long. When the generator is configured to output a batchsize of...

Inconsitency in the reported results [email protected] and [email protected]:0.95

3

Hi, I use the latest version of the repo. There is a huge inconsistency between the reported [email protected] and [email protected]:0.95 as printed during the training versus the results.png E.g. during...

Wrong kpts in train batch

3

Hi, I sometimes have wrong keypoints in a training batch when training on the included coco_kpts.yaml file and default labels as downloaded from the repo, see attached img.  I...

Unclear usage of training look and training_steps in ActivationAPoZRankPruner (model compression, pruning)

4

**Describe the issue**: The workings of the ActivationAPoZRankPruner and its example [examples/model_compress/pruning/activation_pruning_torch.py](https://github.com/microsoft/nni/blob/bfacbc28de70fa30d24676e6dc13dae587ab181e/examples/model_compress/pruning/activation_pruning_torch.py)](https://github.com/microsoft/nni/blob/bfacbc28de70fa30d24676e6dc13dae587ab181e/examples/model_compress/pruning/activation_pruning_torch.py) are unclear. I understand from the [reference paper](https://arxiv.org/abs/1607.03250) a pre-train step is required (if needed), and after pruning...

Hi, How can I calculate or get the values for the `bbox.txt` for the NSVF dataset? According to [NSVF repo itself](https://github.com/facebookresearch/NSVF#dataset) where the bbox.txt file contains a line describing the...

I would love to add simple training visualization like in the [torch-ngp repo](https://github.com/ashawkey/torch-ngp#gallery--update-logs). They fixed this nicely in a small loop here: https://github.com/ashawkey/torch-ngp/blob/ccc703001720119e6b926c6d6971e0433c492b89/nerf/gui.py#L429 I would have fixed this in a...

Would it be possible to run the appxbundle in the background on the hololens?

5

Currently, this is not possible. Is this possible, and what would be the best way to do this? I want to show a Unity application in the hololens 2 (the...

Pointers for TensorRT model with uint8/int8 input

19

By using [pytorch-quantization](https://docs.nvidia.com/deeplearning/tensorrt/pytorch-quantization-toolkit/docs/index.html) i was able to create TensorRT engine models that are (almost) fully int8 and have lower latencies than FP16 equivalent models. One of the downsides is that...