kai

Error when preprocess test dataset

2

When I use the command "python lib/core/data_preprocessor.py --cfg configs/kitti/3dssd/3dssd.yaml --split testing --img_list test", I got the following error. AttributeError: 'KittiDataset' object has no attribute 'mixup_db_class' With debugging, I found that...

Cannot reproduce result for coco

4

Following readme construction, I only got AP 0.065 after training on Seed 1 Percent 1 Coco standard while 25.81 ± 0.28 (+16.76) is claimed in the paper. Here's my setting:...

SD3 cannot finetunes a better model (hand and face deformation)?

3

### Describe the bug I want to finetune sd3 to improve its human generation quality with 3million high-quality human datasets (which has been proven useful on sdxl and other models)....

qwen2.5vl train loss is a lot lower than qwen2vl but performance is wired

4

I use the same codebase transformers train qwen2.5vl and qwen2vl. For qwen2vl the loss is normal around 2.0, but for qwen2.5vl the loss is super low around 0.4. But the...

Low loss but highly disorded output

2

After finetuning on qwen2.5vl , I got a very low train and validation loss around 0.67, but the output is disordered and worse than before. Could you share some experiences...

求一份数据，已经申请还没得到回复

3

感谢非常优秀的工作，可以发一份数据给我吗 [email protected]