Gavia Gray

It's not just the warning, that `if/else` block also sets `num_betas=10` whenever the warning is printed. Later this is used to slice the beta parameters. Is that the intended behaviour...

I've managed to use this code (after making some edits in [my fork](https://github.com/gngdb/smplx)) to convert between SMPL-H and SMPL-X using the AMASS sample data. It seems to have worked. My...

I figured out one problem with it. I wasn't unpacking the data to plot correctly and I was leaving in the root orientation when slicing out the body pose. Fixing...

I've managed to plot sequences of frames using the code from [human_body_prior][hbp]. But, for the same frames, passing them through the `smplx` appears to plot the same frame every time...

If you can load the data into [fairmotion](https://github.com/facebookresearch/fairmotion) you can save to `.bvh` with their [bvh.save](https://github.com/facebookresearch/fairmotion#data-saving). AMASS files can be loaded using `fairmotion.data.amass`. I'm not an author of this repo...

Those betas were before fixing the problem described in #113, after they look like this: ``` betas: -0.916 +/- 0.014 -0.921 +/- 0.026 -0.171 +/- 0.051 -1.407 +/- 0.040 0.889...

Oh, actually those SMPL-H files are already `.npz` obviously `models/smplh_raw/female/model.npz`. Does that mean they don't need to be merged with this tool?

The error above comes from trying to open a `.npz` file as a `.pkl`. It seems like they do still need to be merged however. I was able to get...

Also, if it's supposed to be an `.npz` file, why is it listed as a `.pkl` file in this section of the wiki? https://github.com/vchoutas/smplx/blob/master/README.md#model-loading

I also don't understand how the `transfer_model` is supposed to work. I found this slide in the [SMPL-ify presentation](https://files.is.tue.mpg.de/black/talks/SMPL-made-simple-FAQs.pdf) that suggests it's easy to save to `.obj` files but even...