You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
the link said checkpoint have been browsed and downloaded too many times, cannot download now
I have found before that this is usually if there is high traffic plus not being logged into google. It's not the best workaround, but at least to get started if you are logged into a google account then the downloads may work fine even with high traffic. It's a mystery what the real limit is on drive.
As one more model hosting alternative for you to consider, HuggingFace has been doing a good job of providing a free home for neural network weights. There's file versioning and a readme, and no restrictions on downloading, no account requirement for downloading, etc. CLIP is a good example. This is nice because you can use those links directly, like in your python scripts for bootstrapping.
-> https://huggingface.co/new
Just another option for you to think about. Anyway thank you 😄
Activity
lkeab commentedon Jun 12, 2023
Thanks. we will fix the problem soon.
lkeab commentedon Jun 12, 2023
we provide other google drive links:
vit_b: ViT-B HQ-SAM model.vit_l: ViT-L HQ-SAM model.vit_h: ViT-H HQ-SAM model.vit_tiny: Light HQ-SAM modelkjerk commentedon Jun 12, 2023
I have found before that this is usually if there is high traffic plus not being logged into google. It's not the best workaround, but at least to get started if you are logged into a google account then the downloads may work fine even with high traffic. It's a mystery what the real limit is on drive.
Thank you for releasing the code and weights! 🙌
As one more model hosting alternative for you to consider, HuggingFace has been doing a good job of providing a free home for neural network weights. There's file versioning and a readme, and no restrictions on downloading, no account requirement for downloading, etc. CLIP is a good example. This is nice because you can use those links directly, like in your python scripts for bootstrapping.
-> https://huggingface.co/new
Just another option for you to think about. Anyway thank you 😄
lkeab commentedon Jun 12, 2023
Thanks. We will consider the hugging-face storage.
lkeab commentedon Jun 13, 2023
Hi, the hugging-face model storage is ready now.