Feichi Han

PaddleHub: 1.7.1 PaddlePaddle: 2.1.0 系统: Windows11 python版本: 3.7 代码: ``` import paddlehub as hub module = hub.Module(name="pyramidbox_lite_mobile_mask") import cv2 capture = cv2.VideoCapture(1) while True: ret, frame = capture.read() if ret...

目标检测模型输出结果如何用Python解析

6

我在树莓派4B上部署了一个用PaddleDetection训练的目标检测模型。 训练时模型配置文件选用pp-yolo+mobilenetv3_small, 最终在树莓派里用Python+paddlelite的方式部署, 运行能够成功, 但是输出的结果是一系列六元组, 不知道该如何利用这些信息在图片中画框, 希望得到画框的具体坐标. Paddlelite == 2.11 PaddleDetection == develop 具体输出如下： [[0. 0.07142854 0.11334497 0.7792644 0. 0. ] [0. 0.06879957 0.53201455 0.10974024 0. 0. ] [0. 0.0354831...

### 问题确认 Search before asking - [X] 我已经搜索过问题，但是没有找到解答。I have searched the question and found no related answer. ### 请提出你的问题 Please ask your question 请问Picodet训练完毕后, 我的模型结构在最终输出时有8个输出部分, 请问这个些数据如何利用python解析出来并完成图片检测框的绘制呢? 

PPHuman可以增加自定义视频的行为检测教程吗?

6

目前来看, https://github.com/PaddlePaddle/PaddleVideo/tree/develop/applications/PPHuman#readme 根据这个教程走自己录制的视频会出现一些问题, 希望官方可以给出关于实际增加自定义动作到数据集的步骤, 不然class_id和label如何修改目前个人感觉很困难

PPHuman训练出现错误

2

当进行到PPHuman教程的训练阶段时: `python main.py -c applications/PPHuman/configs/stgcn_pphuman.yaml` 会出现如下报错: ``` Traceback (most recent call last): File "main.py", line 139, in main() File "main.py", line 135, in main profiler_options=args.profiler_options) File "/home/aistudio/PaddleVideo-develop/paddlevideo/tasks/train.py", line 272, in...

About QBatchNormalization is not support QKeras po2 quantizer

1

## Prerequisites Please make sure to check off these prerequisites before submitting a bug report. - [Y] Test that the bug appears on the current version of the master branch....

Keras Reshape Layer is Built with Error

1

## Quick summary I want to synthesize a model with reshape and concatenate layers through hls4ml, but when I build the generated HLS file, there will be a problem of...