Said Taghadouini
Said Taghadouini
Thank you for your quick reply! I have tried the second way by providing different prompts(text-to-image and text-to-image-with-template) but the results were not what I was looking for(the handwriting didn't...
Any updates on this?
thanks for the prompt response 😄 I totally understand the need for quick development. did you try any large scale training on AMD? i don't know if that's the goal...
it requires adding the specifics of siglip in the model's forward. The existing code can easily be modified to do that
Only the original ViT is supported! The openai ViT is different
Hey @kirklandsign, Did you set the NDK_ROOT_PATH variable either in the env or in the CMakeLists.txt : `set(CMAKE_ANDROID_NDK /path/to/android-ndk-r26b-linux)` after downloading the NDK?
Great to hear! Are you running this on a Phone? I'm interested to know what is the setup
It is possible : refer to the [GGML](https://github.com/ggerganov/ggml?tab=readme-ov-file#using-cublas) library.
I can add it if that's something you need. One issue is that i do not have a local GPU so it will be hard to test. would you like...
no plan for supporting it. should be easy to adapt the code though