Understanding generated output format and saving output to binvox file.

[Imperial-Dragon]

Greetings,
I was trying to save the generated 3D volume in a binvox file. I followed your suggestions provided in issue #11 and have following questions :

1. In test.py, on line 149,
   gv = generated_volume.cpu().numpy()
   this gv is a numpy array of shape (1, 32, 32, 32). and the sample output of gv is as follows

How does these values represent our generated 3D volume ?

2. As per your suggestions in issue #11, to save the generated output as a binvox file, I added following code snippet :
   
   But, all saved binvox files are of size 310 bytes only.
   To solve this problem you suggested to convert np.array to np.bool before saving, but I did not understand how to do that as don't know what the array value represents.
   Can you help me to convert this np.array to np.bool ?

[hzxie]

These values are probabilities. You can simply convert them to booleans by:

threshold = 0.4
gv = gv > threshold

[Imperial-Dragon]

Thanks :)