-
执行
` python tools/demo.py image -f exps/ppyoloe/default/ppyoloe_s.py -c model_data/ppyoloe_s.pth --path assets/dog.jpg --conf 0.15 --tsize 640 --save_result --device gpu`
遇到问题
![image](https://user…
-
**PaddleDetection team appreciate any suggestion or problem you delivered~**
## Checklist:
1. 查找历史相关issue寻求解答/I have searched related issues but cannot get the expected help.
2. 翻阅[FAQ](https:/…
-
前陣子有在研究Megvii的YOLOX,十分感謝您將PPYOLOE implement在同樣平台上。
由於deployment的平台不支持hard sigmoid,因此嘗試換成sigmoid後再finetune。
目前使用您提供的PPYOLOE-s pretrained weight去finetune,同樣是訓練在COCO資料集。訓練時遇到幾個問題,有些未解決的部份再請您解惑一下,謝謝!
…
-
环境:
docker: paddlepaddle/paddle:1.7.1-gpu-cuda9.0-cudnn7
paddledetection: release0.2
模型:yolov3_r50vd_dcn.yml,yolov3_r34_voc.yml
显卡:P4, V100
固定config中的batch_size,采用2卡训练:
![image](https://user-i…
-
### 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. 多类别目标跟踪使用MCFairMOT 训练bdd100k数据集的格式应该如何准备?[文档…
-
### bug描述 Describe the Bug
Picodet INT8 is slower than FP32 when inference with MKLDNN.
CPU:Intel(R) Xeon(R) Gold 6271C CPU @ 2.60GHz
thread num:8
FP32:3.09 s
INT8:3.13 s
My model and scri…
-
查看了下faster rcnn的训练配置,几乎都是用的“faster_fpn_reader.yml”文件配置的,但是batchsize=1,base_lr=0.01,paddledetection项目里也没有对应的训练建议,想问下如果在多gpu下训练faster rcnn模型也是采用类似于下面ppyoloe一样的脚本去跑么。
```
python -m paddle.distributed.l…
-
## 描述问题/Describe the bug
A clear and concise description of what the bug is.
1 picodet和ppyoloe关于检测头head有没有类似于下图(YOLOX)这样的图解?如何与前面的csp-neck或者lc-neck连接?
![image](https://user-images.githubuserconten…
-
### bug描述 Describe the Bug
训练ppyoloe时报错,简单的复现了一下报了同样的错误
import paddle
import paddle.nn.functional as F
a = paddle.ones((32, 8505, 4, 17))
b = paddle.to_tensor([0. , 1. , 2. , 3. , 4. , 5. , 6. …
gjghs updated
2 years ago
-
### 问题确认 Search before asking
- [X] 我已经查询[历史issue](https://github.com/PaddlePaddle/PaddleDetection/issues),没有发现相似的bug。I have searched the [issues](https://github.com/PaddlePaddle/PaddleDetection/is…