huxian0402

Hi, Here is full code snippet. def evaluate_model(model, int): print("evaluate_model start...") model.eval() num_test = 0 nmes_merge = 0.0 test_data = data.facepp_st_data('test') for batch in test_data: img = batch['image'] lms_gt =...

Thanks for your reply, I dump all the encodings generated after the sim.compute_encodings step and find that all the encodings are valid. And I find another strange problem. I tried...

Thanks， Yes ，I visualize the offset with the script from original implementation ， and the offset is got from your other repo : https://github.com/Zardinality/TF_Deformable_Net . I have successfully run your...

@SlongLiu When I want to draw a picture using LLaVA-Plus, for example by inputting "Draw a picture with a woman holding a dog by a leash" to the LLaVA-Plus Chatbot,...

> I added a new branch `fast_image_sizes`. Could you please test with the branch? @kohya-ss @RefractAI > I added a new branch `fast_image_sizes`. Could you please test with the branch?...

> I don't think there has been a report like this before. If you restart it, caching may start from where it left off. > > Currently, it is slow...

> I don't think there has been a report like this before. If you restart it, caching may start from where it left off. > > Currently, it is slow...

Hi @Bluear7878 @lmxyy This PR works great — LoRA v2 and Qwen-Image have both been tested and confirmed by several users. Everyone’s really looking forward to official support! 🙏 Any...

> Thank you all for the interest in this feature. I am planning to begin development on this feature support soon. I will share my progress here. @Bluear7878 Hello, may...

> > > > > I am done. It needs a review, but it works fine. > > > > > I plan to test and add the QWen-image edit...