Comments (18)
Linaom1214
commented on May 14, 2024
I train a single class model by yolov7 and conver trt. python3 yolov7/export.py --grid --simplify python3 T_F_Y_S/export.py -o xxx.onnx -e xxx.engine -p fp16 --end2end when I use trt model inference, the boxes and class are same as pth model, but conf is negative and always in [-0.4,-0.6] can you help me,thanks!!!
provide your model file (onnx model)
from tensorrt-for-yolo-series.
lxzatwowone1
commented on May 14, 2024
I train a single class model by yolov7 and conver trt. python3 yolov7/export.py --grid --simplify python3 T_F_Y_S/export.py -o xxx.onnx -e xxx.engine -p fp16 --end2end when I use trt model inference, the boxes and class are same as pth model, but conf is negative and always in [-0.4,-0.6] can you help me,thanks!!!
prevoid your model file (onnx model)
好的，感谢，您意思是我转onnx时候有问题么，prevoid 这个单词我没明白，看您地址是西安，索性我就用中文了TT
from tensorrt-for-yolo-series.
Linaom1214
commented on May 14, 2024
I train a single class model by yolov7 and conver trt. python3 yolov7/export.py --grid --simplify python3 T_F_Y_S/export.py -o xxx.onnx -e xxx.engine -p fp16 --end2end when I use trt model inference, the boxes and class are same as pth model, but conf is negative and always in [-0.4,-0.6] can you help me,thanks!!!
prevoid your model file (onnx model)
好的，感谢，您意思是我转onnx时候有问题么，prevoid 这个单词我没明白，看您地址是西安，索性我就用中文了TT
提供一下您的模型文件， 我看一下 [email protected]
from tensorrt-for-yolo-series.
lxzatwowone1
commented on May 14, 2024
我大概找到了问题，pth转onnx的时候，也就是
python3 yolov7/export.py --grid --simplify这一步后 onnxruntime推理输出的置信度score已经不对了，看起来好像都减半了但是细看也不是减半，但是用您的代码转也并没有得到和onnx同样的置信度结果，百思不得其解TT
from tensorrt-for-yolo-series.
lxzatwowone1
commented on May 14, 2024
新的发现！
zhe'bu'f这部分代码貌似有问题，我在尽力修改，稍有成效
from tensorrt-for-yolo-series.
Linaom1214
commented on May 14, 2024
end2end 和 不包含 nms的模型差距很大吗？ 之前我测试的是好的
from tensorrt-for-yolo-series.
lxzatwowone1
commented on May 14, 2024
您是用几类模型测的
from tensorrt-for-yolo-series.
Linaom1214
commented on May 14, 2024
您是用几类模型测的
80 类 coco
from tensorrt-for-yolo-series.
lxzatwowone1
commented on May 14, 2024
官方那个不行的
from tensorrt-for-yolo-series.
Linaom1214
commented on May 14, 2024
官方那个不行的
????
from tensorrt-for-yolo-series.
Linaom1214
commented on May 14, 2024
我重新测试了一次，多目标、单目标 代码都是好的呀，端到端和不包含nms的模型是可以对齐的， 也许是多Batch的原因，我能想到的就只有这了，等我搞完论文好好研究一下。
from tensorrt-for-yolo-series.
lxzatwowone1
commented on May 14, 2024
感谢 我这边还在尝试
from tensorrt-for-yolo-series.
Linaom1214
commented on May 14, 2024
感谢 我这边还在尝试
确实是 多batch的原因 我在更新动态batch的时候，发现转换会出错， 您 那边有进展吗
from tensorrt-for-yolo-series.
ZhuYoukang320
commented on May 14, 2024
我也遇到分数为负数的情况，检查了很多次代码都没问题。后来偶然发现是FP16时才会出现负的分数，FP32结果是正常的
from tensorrt-for-yolo-series.
Linaom1214
commented on May 14, 2024
我也遇到分数为负数的情况，检查了很多次代码都没问题。后来偶然发现是FP16时才会出现负的分数，FP32结果是正常的
FP16 会有一定的精度损失
from tensorrt-for-yolo-series.
jiguanglu
commented on May 14, 2024
fp16的score与原始pt推理的score的和基本上是1，很奇怪。
from tensorrt-for-yolo-series.
EdisionWew
commented on May 14, 2024
我也遇到分数为负数的情况，检查了很多次代码都没问题。后来偶然发现是FP16时才会出现负的分数，FP32结果是正常的
你这个问题解决了吗，我这边也遇到了非负的情况，可以交流下吗
from tensorrt-for-yolo-series.
ZhuYoukang320
commented on May 14, 2024
我也遇到分数为负数的情况，检查了很多次代码都没问题。后来偶然发现是FP16时才会出现负的分数，FP32结果是正常的
你这个问题解决了吗，我这边也遇到了非负的情况，可以交流下吗
没有解决呢，可能是tensorrt nms插件的问题吧。FP16的话我最后是放弃了端对端nms的方式，自己做后处理
from tensorrt-for-yolo-series.
Related Issues (20)
- YOLOv7 Tensorrt converted model inference is equal to PyTorch model HOT 3
- int8 vs fp16 加速倍数能有多少？ HOT 1
- How to use engine in a process or a thread HOT 4
- how to deploy in multiple nvidia card, such as a computer with 8 3060 card?
- Add dynamic batch support for converting from onnx to .engine?
- auto in_dims = engine->getBindingDimensions(engine->getBindingIndex("image_arrays")); HOT 1
- En715 Jetson xaiver Nx Yolov7.trt Not detect HOT 2
- yolov7，official，int8，onnx-> trt报错 HOT 3
- c++ endtoend 关于预测的置信度绘制 HOT 4
- memory leak: Destroy function does not work
- Detection duplicates with fp16 on Jetson Nano (TensorRT v8.2.1.8) HOT 2
- Support for windows?
- License? HOT 4
- 关于V8 tensorrt 出现乱框的情况 HOT 33
- TensorRT Conversion Issue "TypeError: pybind11::init(): factory function returned nullptr" HOT 2
- yolox 自己训练的模型 trt推理 位置不对 HOT 1
- int8量化的时候，输入是多个，怎么修改呢？ calib_shape = [calib_batch_size] + list(inputs[0].shape[1:])不对吧 HOT 4
- Error Code 1: Serialization (Serialization assertion creator failed.Cannot deserialize plugin since corresponding IPluginCreator not found in Plugin Registry) HOT 2
- wrong confidence score (negative confidence score) on Jetson Nano inference HOT 3
- usage example for image_batch.py HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from tensorrt-for-yolo-series.