Closed RuijieZhu94 closed 3 months ago
Hi @RuijieZhu94, thanks for your interest in our work.
Yes, there is a small bug regarding feature extraction due to code cleaning. It is mainly related to (batch, view)
dimension conversion, it does not affect the testing since testing keeps batch_size=1
. We have already corrected it in our last commit (297338f54d74e7beb4ca5e0700dee22090b836a4). We have re-trained the model (after fixing the aforementioned bug) using both single GPU and multi-GPUs configurations, and they both reproduced the results of the released model.
Would you mind updating the code following our last commit (297338f54d74e7beb4ca5e0700dee22090b836a4) and re-training the model? Let us keep this commit open for you to update the results. For a quicker debugging process, your model should reach around PSNR=23 at step 10K with the updated code, which is around PSNR=20 at step 10K if it contains the aforementioned feature extraction bug.
By the way, we use batch_size=14
by default (a smaller batch_size might slightly harm the performance but should not be that much). And the LPIPS weight is 0.05; the lr scheduler is 1cycle with lr=2.e-4
, as we have updated in this commit (660f49cb7127166af6221f1df2c0d09606b56270). Make sure you have also synchronised your code base (if you have made any changes) with the aforementioned commits.
Hi Yuedong, thanks for your prompt reply, I will retrain this model in the next few days.
Hi Yuedong, I retrained this model with bs=12, and got the result:
psnr 26.31555430801481
ssim 0.8676635705885196
lpips 0.12932708359573464
Thank you for your help.
@RuijieZhu94 Could you share the link of the training dataset? I reach out the author of the pixelsplat for link. However, i can not open the link.
@RuijieZhu94 Could you share the link of the training dataset? I reach out the author of the pixelsplat for link. However, i can not open the link.
Please contact me by email: ruijiezhu@mail.ustc.edu.cn.
Hi Yuedong, thank you for open source your great work!
When I trained the model using 3 Nvidia RTX 3090s (batch size 4 per GPU), I got significantly worse results on the re10k.
Will fewer batchsize or multi-GPU training significantly affect the performance of the model? By the way, I use the official weights and can get results consistent with the paper.