Tao Huang
Tao Huang
@WalkerPan 没有解决，升到mojave之后就没有尝试过了
Hi @dydxdt , 模型文件是用pytorch训练及存储的，后缀名不会影响文件内容。LightViT更关注FLOPs、inference latency和精度，不注重减小模型参数量。
Hi @kunsaram01 , It seems like your training can be successfully launched with some ignorable warnings. The `mc` module is an internal extension in our training environment, and it is...
Dear @songyang86 , Thanks for your detailed implementations. Our code of semantic segmentation is based on the code in [CIRKD](https://github.com/winycg/CIRKD), so we directly report the mIoU in CIRKD paper, and...
Hi @songyang86 , I've trained the DIST on deeplabv3 R101-R18 setting, and got a highest validation mIoU of 77.24. You can see the log below. [deeplabv3_resnet101_resnet18_log.txt](https://github.com/hunto/DIST_KD/files/11541380/deeplabv3_resnet101_resnet18_log.txt)
哈喽，可以提供下你的任务细节吗？是论文中的任务还是其它任务？用的什么模型呢？
你试着把 `${}` 去掉看看呢。用下面的命令： sh tools/dist_train.sh 1 configs/strategies/distill/dist_cifar.yaml cifar_resnet20 --teacher-model cifar_resnet56 --experiment checkpoint --teacher-ckpt ./ckpt/ckpt_epoch_240.pth
估计是你的sh版本的问题，你可以用 `readlink -f $(which sh)` 看一下是什么版本。 或者，使用`bash tools/dist_train.sh`或`./tools/dist_train.sh`试试。
我猜想，你把最新命令里的--nproc_per_node=2改为--nproc_per_node=1应该就可以了。你之前端口冲突了，需要用--master_port=25641修改端口
Hi @JINzezhong7 , You can see the code below. For logits with shape [B, C], we reshape it to [B, C, 1, 1], and use 1x1 convolutions in diffusion loss...