Comments (19)
WongKinYiu
commented on May 27, 2024
resume training is not yet supported.
from pytorch_yolov4.
xiaozhangya
commented on May 27, 2024
resume training is not yet supported.
现在支持了吗？ 加载模型训练重新训练？
from pytorch_yolov4.
WongKinYiu
commented on May 27, 2024
在這裡加上ema.updates = start_epoch * nb // accumulate就可以
from pytorch_yolov4.
xiaozhangya
commented on May 27, 2024
在這裡加上
ema.updates = start_epoch * nb // accumulate就可以
这个您试过吗？ 原理什么呢？另外有u3版本的加载位置吗？ 谢谢
from pytorch_yolov4.
xiaozhangya
commented on May 27, 2024
在這裡加上
ema.updates = start_epoch * nb // accumulate就可以
下面两张图 就是使用您说的方法重新加载模型，但是随着训练继续进行 各项指标都在下降，很疑惑。模型能支持再次训练吗？这是一个很大的问题，做实验等于白做了
from pytorch_yolov4.
WongKinYiu
commented on May 27, 2024
u3版還沒時間整理新的程式碼
from pytorch_yolov4.
xiaozhangya
commented on May 27, 2024
u3版還沒時間整理新的程式碼
之前我问过您 说的是u3、u5都是可用，我最近在做实验，发现了好多bug! 每次训练都会浪费很长时间呢。您能尽快解决u3这个bug吗？学了这么长时间了 也不想再去另换其他版本的代码了。急求！麻烦您了
from pytorch_yolov4.
xiaozhangya
commented on May 27, 2024
保存时giou=1.4，map=0.47，再训练一个epoch变2，map变0.3
现在还在用这个版本的代码吗？
from pytorch_yolov4.
xiaozhangya
commented on May 27, 2024
u3版還沒時間整理新的程式碼
那u5版本加上那行确实可以重新加载保存的模型吗？ 如果可以就重新换这个版本学习了。
from pytorch_yolov4.
WongKinYiu
commented on May 27, 2024
可以呀
u3應該是加在192行 您可以測試看看
from pytorch_yolov4.
xiaozhangya
commented on May 27, 2024
可以呀
u3應該是加在192行 您可以測試
u3版本就是加载到这一行呀
那这样都不行 u5可行吗 现在没有多余的时间去整理这些原始模型上的bug了。
from pytorch_yolov4.
WongKinYiu
commented on May 27, 2024
u5不是preview版 已測試可行 (u5_preview, u5_mish_preview未測試)
u3目前release的是preview版 我這本機已經改版 所以沒有測試過舊版
from pytorch_yolov4.
WongKinYiu
commented on May 27, 2024
因為不確定您的訓練流程
建議您把這裡先註解掉
from pytorch_yolov4.
xiaozhangya
commented on May 27, 2024
u5不是preview版 已測試可行 (u5_preview, u5_mish_preview未測試)
u3目前release的是preview版 我這本機已經改版 所以沒有測試過舊版
您好 我使用是这个版本的u3_preview的yolov4
from pytorch_yolov4.
xiaozhangya
commented on May 27, 2024
u5不是preview版 已測試可行 (u5_preview, u5_mish_preview未測試)
u3目前release的是preview版 我這本機已經改版 所以沒有測試過舊版您好 我使用是这个版本的u3_preview的yolov4
到底u3版本和u5版本分别是那个可以使用？您如果更新了 能分享到GitHub吗？ 另外刚才您说的注释，还是基于u5的，u3那一段代码不会被执行的。所以跟那一段代码似乎没有太大关系
from pytorch_yolov4.
xiaozhangya
commented on May 27, 2024
因為不確定您的訓練流程
建議您把這裡先註解掉
到底u3版本和u5版本分别是那个可以使用？您如果更新了 能分享到GitHub吗？ 另外刚才您说的注释，还是基于u5的，u3那一段代码不会被执行的。所以跟那一段代码似乎没有太大关系
from pytorch_yolov4.
xiaozhangya
commented on May 27, 2024
因為不確定您的訓練流程
建議您把這裡先註解掉
还在吗？ u3版本和u5版本分别是那个可以使用？
from pytorch_yolov4.
xiaozhangya
commented on May 27, 2024
u3目前release的是preview版 我這本機已經改版 所以沒有測試過舊版
u3目前release的是preview版 我這本機已經改版 所以沒有測試過舊版，您是有最新的u3版本可以使用吗？能分享到github吗？谢谢啦！！
from pytorch_yolov4.
joe660
commented on May 27, 2024
resume training is not yet supported.
U3_pre版本不支持保存的模型在训练 但是您预训练的权重我却可以拿来直接使用。。
我假如在公开数据集上跑出一个预训练权重，然后再自己数据集上加载这个预训练权重运行可以吗？？？？
谢谢您的回复
from pytorch_yolov4.
Related Issues (20)
- anchor和SAM问题？ HOT 1
- yolov4-tiny P值很低，R值很高，怎么办？
- AP is very low
- High GPU usage
- weights/yolov4-pacsp.pt HOT 1
- Training Error about yolov4-tiny in branch u5
- How to solve this error? RuntimeError: indices should be either on cpu or on the same device as the indexed tensor (cpu) HOT 7
- Training/(YAML) Data Loading Error: train.py line 537 and line 67 HOT 1
- Training Error: TypeError: No loop matching the specified signature and casting was found for ufunc greater HOT 4
- How to implement freeze layer HOT 1
- Weight links are down
- pretrained_ models HOT 1
- 运行detect.py时出现错误，请问该怎么改 HOT 1
- RuntimeError: shape '[2, 3, 9, 80, 80]' is invalid for input of size 230400 HOT 1
- anno_json = glob.glob('../coco/annotations/instances_val*.json')[0] # annotations json IndexError: list index out of range
- Links to weight files are not functioning HOT 2
- test.py运行提示label错误
- RuntimeError: a view of a leaf Variable that requires grad is being used in an in-place operation.
- Error training your own data set HOT 1
- get_coco2017 issue unziping labels inside coco2017labels.zip HOT 1
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 pytorch_yolov4.