weigq

in fact, same as the origin repo, the mapping is not one-to-one (the jaw joint only appears in 3D pose), most 3D pose estimation methods also do as this.

seems you are using a lastest version of pytorch. You can try replace the `loss.data[0]` with `loss.item()`

Good, i will check that.

The original data in my google driver has lost. I am reconstructing this repo and the pre-processed data. I will update asap😀.

@Ssrzbao Did you use Pyhton2 ? The data is saved using torch in python2.7, it will cause error when loading using Python3

@ahwaleed You`d best use 2.7, I have tested on python3, where some errors occur.

@Spandan-Madan I have upload some pretrained checkponints [here](https://drive.google.com/drive/folders/1weT4YaFWm5JOANP1FHWhw7l7In_fdDUn?usp=sharing).

@mcchran ```bash export TERM="xterm-256color" # If you come from bash you might have to change your $PATH. # export PATH=$HOME/bin:/usr/local/bin:$PATH # Path to your oh-my-zsh installation. export ZSH="/home/weigq/.oh-my-zsh" # Set...

 looks good

最近没应该没有分数的变动吧