SiLang comments

Results 7 comments of


                                            SiLang

Weights for the ViT-B-32 model

Hi, Sorry for the late reply. I've sent you the linkage of weights, yes? I will release them recently.

Question concerning OOD detection

Hi, Sorry for the late reply. - "I'm using the .pt weights you kindly provided," The provided weights are for CLIPN with learnable prompts. So these models didn't know the...

Code issue

Hi, Sorry for the late reply. For the zero-shot inference of CLIPN, you should run src/zero_shot_infer.py. We re-define the inference model in /src/classification.py. In short, the text encoder and 'no'...

Having trouble installing requirements on MacOS with M3

Hi Sorry for the late reply. It's not necessary, lib. In fact, when I first worked on CLIPN, I didn't create a separate environment for it but rather a mix...

How can we get the model weights of CLIPN

Hi, You can find the pre-trained model in the Reproduced Results Section of README ![image](https://github.com/user-attachments/assets/b270d0b5-8a8a-4cb0-9ad8-de898bf6bb77)

about "Cloud-init"

Hi, It's not necessary, lib. In fact, when I first worked on CLIPN, I didn't create a separate environment for it but rather a mix of many projects, which led...

How to subdivide the same category, for example, how to distinguish Persian cats, coffee cats, and jingle cats, which are all cat categories?

Hi, If you are working on fine-grained OOD detection, a simple way is to treat the fine-grained categories as distinct, such as "It's a coffee cat" and "It's not a...