Chen94yue

How to generate "final_split_voc_10_shot_instances_train2017.json"

1

I want to use fewx to train my datasets. Could you release the code about how to generate "final_split_voc_10_shot_instances_train2017.json"?

https://github.com/fanq15/FewX/blob/3392c747471caa6f0174748aaa72eaf3db18be91/fewx/data/dataset_mapper.py#L242 https://github.com/fanq15/FewX/blob/3392c747471caa6f0174748aaa72eaf3db18be91/fewx/data/dataset_mapper.py#L215

the prototxt file is shown as fallows: name: "ResNet-50" layer { name: 'input-data' type: 'Python' top: 'data' top: 'im_info' top: 'gt_boxes' python_param { module: 'roi_data_layer.layer' layer: 'RoIDataLayer' param_str: "'num_classes': 2"...

The MR is not as good as that in the paper.

9

I have trained the proposed model on my server with P40 GPU. The MR on Cityperson val dataset is 18% with one GPU (16% for two). Where might the problem...

腾讯视频越下越慢 使用-cs无效

2

下载腾讯视频越下越慢，使用nethogs查看实际annie占用网络只有1kB/s，然而显示速度很快。已经带上-cs 10了，使用-cs 5也一样。

我注意到在生成label的时候只使用了kinect的标注数据：https://github.com/GouMinghao/rgb_matters/blob/2e176728f6dae5e878447240ce6a2423bf770f36/rgbd_graspnet/data/utils/gen_label.py#L378 但是我使用 ``` assert grasp_labels == grasp_labels_rs ``` 检查两个标签是否一致的时候发现并不一致，这样在训练rs的模型的时候不会产生问题吗？

使用Alibaba-NLP/gte-Qwen1.5-7B-instruct模型报如下错误，环境为官方提供的最新版本镜像。 Traceback (most recent call last): File "/opt/conda/lib/python3.10/site-packages/xinference/api/restful_api.py", line 731, in launch_model model_uid = await (await self._get_supervisor_ref()).launch_builtin_model( File "/opt/conda/lib/python3.10/site-packages/xoscar/backends/context.py", line 227, in send return self._process_result_message(result) File "/opt/conda/lib/python3.10/site-packages/xoscar/backends/context.py", line 102, in...

API调用问题

2

我使用api的时候碰到这个问题: ``` [2023-09-13 19:47:17,004] [INFO] [real_accelerator.py:158:get_accelerator] Setting ds_accelerator to cuda (auto detect) [2023-09-13 19:47:22,544] [INFO] building VisualGLMModel model ... [2023-09-13 19:47:22,561] [INFO] [RANK 0] > initializing model parallel with size...

在一些私有化部署的场景，出于安全的考虑，想要关闭Metrics，请问该如何操作。