Open themichaelminer opened 1 year ago
Can you try with
!pip install face-alignment==1.3.4
It's saying this after changing the code to that
This error may be due to not downloading the correct model. Previously, I had stored the model on Tsinghua Cloud, but since I have graduated from Tsinghua University, I cannot continue to use it, and the model was deleted. However, I have now uploaded the model to Yandex and it can be downloaded using the following commands:
!mkdir checkpoints
!pip3 install wldhx.yadisk-direct
!curl -L $(yadisk-direct https://disk.yandex.com/d/i08z-kCuDGLuYA) -o checkpoints/vox.pth.tar
that fixed it! awesome!
unable to download
Traceback (most recent call last):
File "/usr/local/bin/yadisk-direct", line 8, in
!mkdir checkpoints !wget -c https://cloud.tsinghua.edu.cn/f/da8d61d012014b12a9e4/?dl=1 -O checkpoints/vox.pth.tar
Change it to this it should work .
I am still getting same error as @themichaelminer,
When I performed what @sarique2003 mentioned I am below getting error, though file is getting created
Even after performing what @yoyo-nb suggested I am getting the same error as @allthingssecurity is getting.
Can anyone please help here.
My bad it has been removed from tsinghua cloud, intially i followed what @yoyo-nb said and it worked , now i am facing the same error again.
@sarique2003 @allthingssecurity @lokesh005 Sorry, this is because Yandex is displaying the message "Download limit exceeded".
According to the Yandex Disk website:
If a public file was downloaded many times during the day, the download limit is activated, and the download may be restricted for a day. If you don't want to wait, save the file to your Yandex Disk using the Save to Yandex Disk button.
You can try manually downloading the model from Yandex or Google Drive, and then upload it to the corresponding location in Google Colab.
I registered links to the models (the author fixed them in the code), manually uploaded all the models to the Checkpoints folder. But it still gives an error .. What could be?
thanks @yoyo-nb it worked