Nicole Finnie

> I am eager to learn how the solution to this problem is approached. Can you tell me where the weights are located and how they are downloaded? I am...

I have the same question, however, see Edward's comment [here](https://github.com/microsoft/LoRA/issues/98#issuecomment-1628825726) To my understanding, Edward resets the parameters of the weight matrix the same way as pytorch's `Embedding.reset_parameters()` ```python # the...

> Yes. It's important that we have either A or B set to zero so the first forward pass doesn't deviate from the original model. It's also important that A...

@KaleidoZhouYN `scale_factor` doesn't really matter in my experiments (I've tried 30~60 after having read your suggestion), where I have a large number of classes. If your loss function is cross...

@himanshurawlani I don't think removing x86 is a solution, since this is intended for x86-based Android devices. You can see the list of ABIs for different CPUs here. https://developer.android.com/ndk/guides/abis.html

@himanshurawlani I see, thank you for letting me know. tensorflow 1.5 has come out recently, you may want to pick up the newest libraries if you want to try. I...