马浩伟

LOSS可视化（LOSS visualization）

1

这里想问一下大家我们应该如何把训练过程中的LOSS参数可视化呢？ Here I would like to ask everyone how should we visualize the LOSS parameters during training?

Sorry,I have just watched the model structure,would you mind telling me the difference between CrossFormer and swin_transformer?

bruteForceKnn is deprecated; call bfKnn instead Faiss assertion 'err__ == cudaSuccess' failed in void faiss::gpu::runL2Norm(faiss::gpu::Tensor&, bool, faiss::gpu::Tensor&, bool, cudaStream_t) [with T = float; TVec = float4; IndexType = int; cudaStream_t...

Dukemtmc is unavailable

3

 I have download the dataset,but it's unavailable

根据实验结果的对比我们可以观察到deeplab在细粒度和粗粒度上都比HRNet要好，为什么要使用HRNet作为车辆部件解析的模型呢？

如题

token不应该是使用的时候动态申请的吗？文档里面说自己设定的是什么意思？求解答一下