DoItWithMe

It's same error even I cast the 4 arguments to int.  > could you double check that all 4 int arguments are indeed ints?

确实，我测试的时候快速重复推了200路的0.5Mbps左右的srt-ts流到srs上，srs的内存占用从开始的不到1g，10几分钟后涨到4.5g

Thanks for your reply, I will provide a minimal reproducible code later

``` import torchvision.models as models import torch import torch.nn as nn from pytorch_metric_learning import losses from loguru import logger as log def main(): init_lr: float = 0.1 batch_size: int =...

Version Info python 3.10.0 torch 2.6.0 torchaudio 2.6.0 torchvision 0.21.0 pytorch-metric-learning 2.8.1