Ahmed Wasfy

ahmedmwasfy.com [email protected]

Realtor.com Brampton, Ontario

Results 13 comments of


                                            Ahmed Wasfy

Google Recaptcha v3 integration

+1 👍

@auth/express No "exports" main defined

@balazsorban44, can you please advise ?

Ludwig serve <no maximum length is provided>

Facing the same issue, I am trying to fine-tune facebook/opt-350m

Ludwig serve <no maximum length is provided>

@justinxzhao, any thoughts please?

Ludwig serve <no maximum length is provided>

@alexsherstinsky, can you please help?

Ludwig serve <no maximum length is provided>

@alexsherstinsky thank you so much. Let me give that a try and let you know. I appreciate your time and reply

can't import mask_from_lens from monotonic_align

You need to install `pip install git+https://github.com/resemble-ai/monotonic_align.git`

[Question]After fine-tuning, the model performed well, but after merging into GGUF (without quantization), its performance dropped. Why might this happen?

i have been facing the same issue and I found that the solution is to load the base model in 16bit, merge it to the adaptor then save it. After...

[Question]After fine-tuning, the model performed well, but after merging into GGUF (without quantization), its performance dropped. Why might this happen?

@rolandtannous , thank you for your reply. I am going to be running this model on a device, hence the quantization for space. Thanks.

[Question]After fine-tuning, the model performed well, but after merging into GGUF (without quantization), its performance dropped. Why might this happen?

Thank you so much for that incredible insights

1
2
›