Yu Zhu

[Object Detection] Cascade Mask R-CNN

1

**Describe your feature request** Cascade Mask R-CNN **Describe the reference code or paper** Swin detection official code (from mmdet) [here](https://github.com/SwinTransformer/Swin-Transformer-Object-Detection/tree/master/mmdet/models/detectors) **Describe the possible solution** Mask R-CNN is already implemented in...

This implementation is referred to the original pytorch repo. TODOs: - [x] LARS optimizer

**Describe your feature request** Now the dataset for ImageNet2012 requires the "train_list.txt" and "val_list.txt" in the same folder as "train" and "val" which stores the images. Users are required to...

**Describe your feature request** Add recompute for dygraph model training, which aims enlarge the batchsize during the training by free intermediate memories. **Describe the reference code or paper** N/A @jarygrace...

**Describe your feature request** TransGAN: training from scratch **Describe the reference code or paper** official repo: https://github.com/VITA-Group/TransGAN **Describe the possible solution** TODO **Additional context** N/A

**Describe your feature request** Styleformer training from scratch **Describe the reference code or paper** official repo: https://github.com/Jeeseung-Park/Styleformer **Describe the possible solution** TODO **Additional context** Add any other context or screenshots...

**Describe your feature request** Token labeling is used in VOLO model training, implement related classes and methods. **Describe the reference code or paper** 1. The soft target loss and token...

I saw that in the labels with landmark, there are still some faces are not labeled with landmarks, how to handle this case in the training? From the code, it...

opencv error after converted to lmdb

2

Hello, I have converted lmdb using your code without any errors, however when I start train the model, it shows error as following: OpenCV Error: Assertion failed ((scn == 3...