Kadir Nar

Results 229 comments of Kadir Nar

https://github.com/Yujun-Shi/DragDiffusion/pull/6

No. https://anaconda.org/nvidia/cudatoolkit ![image](https://github.com/Yujun-Shi/DragDiffusion/assets/36204372/30c8cd6f-859d-4feb-980f-07e521b8547a)

The conda package only supports version 11.7. If you want to install pip it supports 11.8. But they don't support 12.2 version. ![image](https://github.com/Yujun-Shi/DragDiffusion/assets/36204372/782fd46a-4cee-471f-81fe-fed215b56640)

Can you try this code? ``` pip3 install torch torchvision--index-url https://download.pytorch.org/whl/cu117 ```

Code: ``` outputv0 = pipe( prompt, image=image, mask_image=mask, num_inference_steps=35, guidance_scale=5.0, strength=1.0, num_images_per_prompt=1, cache_interval=5, cache_layer_id=0, cache_block_id=0, uniform=True, output_type='pil', return_dict=True ).images ``` ![image](https://github.com/horseee/DeepCache/assets/36204372/7a096734-c5db-4bdf-8dc1-79135a8dc6b6)

> Hi @kadirnar, please tune the cache_interval, cache_layer_id & cache_block_id. Caching in a deeper layer can improve the quality yet reduce the speedup. Can you suggest sample values? I don't...

I tested it. It does not support Lora models.

@zjysteven I also want to train the flux model with controlnet-tile. How should I prepare the dataset? I created 1024x1024 images using flux-dev. What kind of images should be for...