Koan-Sin Tan
Koan-Sin Tan
CI/CD on Browser-stack is already enabled with github, right now we test One latest phone from each SOC vendor (Pixel, MTK, Qualcomm, Samsung)> Add one more per SOC (latest 2...
FYR. Apple already implemented Stable Diffusion in Swift. Please check if we can reuse Apple's implementation first. https://github.com/apple/ml-stable-diffusion
the Stable Diffusion v1.5 one we use for 4.1 submission (https://huggingface.co/benjamin-paine/stable-diffusion-v1-5)
@AhmedTElthakeb please report number of parameters and FLOPs of the 3 models we use.
For model conversion with AI-Edge-Torch, as I noted at https://github.com/google-ai-edge/ai-edge-torch/issues/269#issuecomment-2505304355 - you don't need GPU - you need a lot of DRAM, which is supposed to be a bug /...
@Mostelk: there is a GPT2 Android tflite app, https://github.com/huggingface/tflite-android-transformers/tree/master/gpt2 @mohitmundhragithub: there is https://github.com/mlcommons/mlperf_client which runs llama 2 7B
@freedomtan to check the GPT2 and the https://github.com/mlcommons/mlperf_client.
check that input format of huggingface tflite models (supposedly, you can simple cast integer token ids to fp16, setting the input tensors, if the expected input is token ids rather...
> check that input format of huggingface tflite models (supposedly, you can simple cast integer token ids to fp16, setting the input tensors, if the expected input is token ids...
> I managed to get input into and output out of the converted model. I used the python version of TFLite but will be moving the code to C++ since...