Experiment details

[KerwenX]

Hello @yccyenchicheng,

I apologize for the inconvenience and hope this message finds you well. I find your research to be highly intriguing and I have some inquiries regarding the experimental details.

Specifically, I would like to inquire about the device(s) used when training the Sdfusion-mm2shape. Additionally, I am curious to know how long it took to complete the training process.

Furthermore, I would like to kindly request if you could consider releasing the training code for Sdfusion-mm2shape as soon as possible.

Thank you very much for your time and consideration. I appreciate any assistance you can provide!

Best regards

[yhhxt]

I have a similar question. I am training a VQ-VAE using DDP framework with 4 NVIDIA GeForce RTX 3090 GPUs. I am using the entire Shapenet dataset, which includes all classes, but the training time is as long as 40,000 hours. I was wondering what dataset you used and how many epochs you ran the model for when you trained a similar model. Also, how long did the training take?Thanks.

[KerwenX]

I have a similar question. I am training a VQ-VAE using DDP framework with 4 NVIDIA GeForce RTX 3090 GPUs. I am using the entire Shapenet dataset, which includes all classes, but the training time is as long as 40,000 hours. I was wondering what dataset you used and how many epochs you ran the model for when you trained a similar model. Also, how long did the training take?Thanks.

Thank you for getting back to me, and I apologize for the delay in my response. However, I have not yet started training on this project, so I am unable to provide an answer to your question at this time.

[yccyenchicheng]

You can monitor through the tensorboard and stops the training if the loss converges or the reconstruction looks decent for you.

It takes around 7-10 days to train the AE, and 10-14 days to train the diffusion models on one GPU. It will converge faster with multiple GPUs with DDP.

[yhhxt]

Thanks a lot!