GDbbq

Results 18 issues of GDbbq

Looking in indexes: https://pypi.tuna.tsinghua.edu.cn/simple Processing ./rknn_toolkit_lite-1.7.3-cp36-cp36m-linux_x86_64.whl Requirement already satisfied: numpy in /home/linjiahui/.local/lib/python3.6/site-packages (from rknn-toolkit-lite==1.7.3) (1.16.3) Requirement already satisfied: psutil in /home/linjiahui/.local/lib/python3.6/site-packages (from rknn-toolkit-lite==1.7.3) (5.9.3) Requirement already satisfied: ruamel.yaml in /usr/local/lib/python3.6/dist-packages...

s$ sudo pip3 install rknn_toolkit_lite-1.7.0-cp36-cp36m-linux_x86_64.whl Keyring is skipped due to an exception: org.freedesktop.DBus.Error.TimedOut: Failed to activate service 'org.freedesktop.secrets': timed out (service_start_timeout=120000ms) WARNING: The directory '/home/linjiahui/.cache/pip' or its parent directory is...

## Checklist: ![image](https://user-images.githubusercontent.com/86715812/231388543-fa80de24-1945-4551-9662-8fbe97822e7c.png) ## 描述问题 使用自己的训练集, python 6.train_ppyolov2_r50_aug_COCO_addneg.py 报错, 运行 python 1.train_ppyolov2_imagenet.py 不会出现错误。。 所以是配置文件的问题? ![image](https://user-images.githubusercontent.com/86715812/231390762-bdfcb3df-ce09-4010-8ea2-32b8183ae91b.png) ## 环境 ![image](https://user-images.githubusercontent.com/86715812/231389451-6148ce26-20df-47bb-8e1e-ec98c9b2a3df.png)

## Checklist: ## 描述问题 ![image](https://user-images.githubusercontent.com/86715812/228502317-377c3a91-a74a-4619-849c-ae4125098bdb.png) ## 复现 1. 跑样例,在AIstudio平台不会报错, 自己的linux下报错(验证时) ## 环境 cuda 10.2 paddlepaddle 最新的 2.4 ## 解决 降低paddlepaddle 2.3.2 ![image](https://user-images.githubusercontent.com/86715812/228502908-2c03a5e6-d324-4456-ae3e-4cc5f206650e.png)

解决办法: https://blog.csdn.net/Thebest_jack/article/details/125649451

Hello, I'm experiencing video artifacts, such as glitching or "garbled" frames, when streaming video using the DeepStream SDK. Attached is a screenshot illustrating the issue I'm encountering. Could you please...

# 设备 unbuntu20 nvidia 1080ti cuda 11.7 ternsorrt 8.6 paddleslim 2.6 paddle 2.6 # 复现过程 1. 下载链接提供yolov5s.onnx, 跑pytorch_yolo_series样例,得到量化后模型 2. onnx推理预测,fp32 和 fp16结果均正常, int8没有框; 3. 运行时,加载tensort模型会打印的消息,可能是造成结果异常的原因;fp32 和 fp16也会打印下面的内容,但是输出的图片结果是正常的。如下图,正常输出的结果 ``` [01/16/2024-03:40:10] [TRT]...

大神您好,我想问下,我从pcn完整点云数据集中,每个类别随机选200个模型,然后加上自己的30个点云模型作为一个类别valve,得到这样一个数据集; 采用adapoint模型进行训练, 其中输入是随机采样后的残缺点云,标签是下采样后的16384个点的数据; 训练过程如下图所示,自己添加的类别数据损失特别大, 模型平均fscore震荡; 这是训练策略问题还是数据问题呢?恳请大神解答,谢谢 ![image](https://github.com/yuxumin/PoinTr/assets/86715812/bf6c710e-419d-4946-bfbe-f226a4abf0c5)

## 环境 - 已安装fastdeploy, 跑其他推理后端都没问题 - 跑https://github.com/PaddlePaddle/FastDeploy/blob/develop/examples/runtime/python/infer_torchscript_poros.py报错 ## 问题日志及出现问题的操作流程 - 【模型跑不通】 ![image](https://github.com/PaddlePaddle/FastDeploy/assets/86715812/150bbc34-5517-40a2-9ef9-9f6068932619)

# 配置 Quadro K620 ![image](https://github.com/PaddlePaddle/FastDeploy/assets/86715812/54c7e5ae-11ca-4d4d-a306-a73e46b25cef) 使用镜像异常 ![image](https://github.com/PaddlePaddle/FastDeploy/assets/86715812/0ae92dc0-4371-4c5f-b75b-99a439b09540)