kzh2ang
kzh2ang
楼主,你好,刚开始入门reid,看了你的代码,最后的评估evaluate_result中的score是算的两个特征矩阵乘积,`score = torch.mm(query_feature, torch.transpose( gallery_feature, 0, 1))`,不是很理解,一个query的feature与gallery的features做度量,为什么要用乘法呀?求指教
你好,今天出现了所有pod正常运行,但是30000端口访问不了的情况,请问该如何排查?  
- kfserving-controller-manager-0 - katib-controller-77675c88df-pswzm - admission-webhook-deployment-6fb9d65887-ntcfc 三个pod都显示CrashLoopBackOff; 分别查看logs后: **[root@iZbp1ho1aruzoda9o2fo6cZ ~]# kubectl logs kfserving-controller-manager-0 -n kubeflow -p** Error from server (BadRequest): a container name must be specified for pod kfserving-controller-manager-0, choose...
你好,基于该仓库,用自己的领域数据训练出来的相似度都是1,是怎么回事?
### Reminder - [X] I have read the README and searched the existing issues. ### Reproduction ### 现象1 用单卡A100 80G,sft qwen1.8的模型,训练的loss一直为0?  ```shell CUDA_VISIBLE_DEVICES=0 python src/train_bash.py \ --stage sft \...
hello, when we have done the train, the `predict.py `can plot the predict boxes, but ,how can we evaluate the `recall` , `precission` and `mAP`
RewardModel计算两个response之间的差异:end_ind的计算是通过`end_ind = max(one_ind, two_ind)`算的,为什么不是直接比较`one_input_ids`和`two_input_ids`差异的最后一个值,也就是`check_divergence[-1]`来获得。
在nodes.py文件中新建了一个节点。 ``` def html_reporter_node(state: State, config: RunnableConfig): """HTML Reporter node that write a html report.""" logger.info(">>>> HTML Reporter write html report") configurable = Configuration.from_runnable_config(config) ... ``` 并在builder.py中创建了reporter到html_reporter的边。 ``` builder.add_edge("reporter","html_reporter") ```...
pnpm dev 报错,node版本23.1.0 Linux系统 --- Last few GCs ---> [58283:0x7e91000] 14451 ms: Scavenge 10230.6 (10239.3) -> 10230.6 (10243.3) MB, pooled: 0 MB, 1.59 / 0.00 ms (average mu = 0.977,...