Train using 128 resolution

[eivillamuelas]

Hello, thanks for the repository

I'm trying to train the model using binvox files with res 128 as you did in your paper.
I used the code python main.py train --tag {your-experiment-tag} -s {path-to-processed-h5-data} -g {gpu-id} but the processing time is the same and I dont see any diferent results compare to res 64.

Is there any parameter I'm missing?

Thanks again

[rundiwu]

Hi,

Did you specify --res 128 when preparing the data? Could you check the processed .h5 file does have voxels of resolution 128? If so, running main.py as above should be good.

[eivillamuelas]

hello thanks yeah it works perfect. I was trying to change the resolution (64, 128, 256) when preparing data but all the final models look similar. Do i need to add something to main.py to change the resolution training? thanksss

[rundiwu]

Just to confirm, are you saying after training on resolution (64, 128, 256), the final generated shapes (of the same resolution) lack diversity (i.e., they look similar)? Or results from res 128 look similar to those from res 64?

In any case, you don't have to add anything to main.py when changing the resolution.