Aria-Zhangjl / E3-FaceNet

[ICML 2024] Fast Text-to-3D-Aware Face Generation and Manipulation via Direct Cross-modal Mapping and Geometric Regularization
15 stars 1 forks source link

Not enough values to unpack #8

Open yzmengnan opened 2 days ago

yzmengnan commented 2 days ago

Hi!

I spent lots of time on trying to run the bash script _run_train_4_E3Face.sh, but all of trials failed.

I think this may resulted by the system platform (win11), or may be it just beacuse that im a green hand here.

Anyway, i turned to run the sample.sh script, but still it failed...

As seeing in the figure, when i tried to run the sample.sh, this issue occurs. Is there any suggestion?

1726571407448

Aria-Zhangjl commented 2 days ago

Hi, All experiments were conducted on a Linux system with RTX 3090 or V100. Therefore, it might be usefule for you to execute the script on a similar device.

Moreover, to run the sample.sh script, please refer to #6 and the readme file for detailed instructions. Specifically, you will need to modify the original clip script to enable the concurrent extraction of text features and token embeddings.

yzmengnan commented 2 days ago

Hi, All experiments were conducted on a Linux system with RTX 3090 or V100. Therefore, it might be usefule for you to execute the script on a similar device.

Moreover, to run the sample.sh script, please refer to #6 and the readme file for detailed instructions. Specifically, you will need to modify the original clip script to enable the concurrent extraction of text features and token embeddings.

get!