ddk-sysu

https://github.com/zimenglan-sysu-512

SYSU ShenZhen, China a little lazy to crazy wechat: zimenglan_sysu oppo内推码，欢迎投递 https://t.oppo.com/1b6IQH

Results 22 comments of


                                            ddk-sysu

where is the rcnn_bboxes come from?

oh get it, just sort the scores of the final ouputs to select top-k bbox

how use the multi-GPU inference ，can you give me some demo code

haha, any update?

Could you please provide more examples to do inference on the different tasks in the paper?

+1

Installation fails

@Frosty774 u should downgrade the torch to 1.x version, e.g. 1.10.2. and then change the cuda path in make.sh file. good luck.

aimV2: Are you planning on releasing the Causal Multimodal Decoder?

+1

Could you please release the UGCrop5K Dataset？Thanks！

any update?

Could you please release the UGCrop5K Dataset？Thanks！

any update?

Full Training from Start：CUDA out of memory.

is it possible to train the q-align using one A100-80G GPU?

nunchaku可以无缝集成吗

+1

Will qwen_image_edit Support Batch Inference and DPO Training?

+1

‹
1
2
3
›