-
### Game or games this happens in
Any game
### What area of the game / PPSSPP
Hey!
The new-ish MSAA option worked fine last time I tried the Libretro core but now it doesn't seem to do any…
-
### 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
有两个问题:
1. (YOLO检测模型建议总batch_size至少大于64去训练,如果…
-
### 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
有两个问题:
1. (YOLO检测模型建议总batch_size至少大于64去…
-
-
### What feature do you want to see added?
After https://github.com/jenkins-infra/update-center2/pull/605, we may no longer have a need for length limitations for `groupId`, `artifactId`.
RPU itse…
-
### 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
我已经可以正常跑通release/2.7分支代码,CUDA安装正常,可以训练SOLOV2实例分…
-
### 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
1.请问下DCNv2算是特殊算子吗,对于使用Paddle Inference推理来说,我看某篇…
-
目前的想法是通过paddle-serving部署。
paddle-serving部署时支持pipeline模式,可以通过定义op来进行多个模型的串联。
但是将模型转换为paddle-serving支持的模型时,目标检测与追踪的模型(pp-yoloe)输出包含两个结果,维度与关键点检测模型(HRNet)的输入并不匹配。并且最终需要通过stgcn来判断是否摔倒,而stgcn需要50帧图像才…
-
### 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
图片太大 20000*20000
切割下来 也有二百多张图
PP-PicoDet模型
每…
-
root@ubuntu:/serving/examples/vision/detection/paddledetection/serving# fastdeployserver --model-repository=./models/
I0906 03:13:30.949881 244 metrics.cc:298] Collecting metrics for GPU 0: NVIDIA Ge…