Comments (8)
from attentionocr.
竞赛数据集没有相似的文档长文本,需要用新的数据集重新训练。
from attentionocr.
我拿公开数据集(art+rects)训练430000次,用来识别车牌,效果也是不太好。前两个字还行,后面完全识别不到。这也是长文本的问题?
结果如下:
['皖', 'K'] [0.7606085 0.9683497 0.95730805]
['皖', 'K'] [0.7606051 0.9683497 0.9573083]
['啵', 'B'] [0.22663411 0.96232164 0.61064464]
['皖', 'G', '▪'] [0.4462814 0.99723685 0.8549502 0.9544695 ]
['皖', 'G', '▪'] [0.4462814 0.99723685 0.8549502 0.9544695 ]
['皖', 'A'] [0.8662637 0.84521544 0.9425101 ]
['啵', 'B'] [0.22663411 0.96232164 0.61064464]
['皖', 'A'] [0.8662637 0.84521544 0.9425101 ]
from attentionocr.
我把新的检测模型放到demo.sceneinception.com,可以先尝试一下效果
from attentionocr.
from attentionocr.
我把新的检测模型放到demo.sceneinception.com,可以先尝试一下效果
您好,感谢您分享这么好的工作,请问有可以finetune的权重吗?docker里的pb好像是不可以训练的?
from attentionocr.
能分享demo.sceneinception.com里的可以finetune的模型吗?非常感谢
from attentionocr.
我拿公开数据集(art+rects)训练430000次,用来识别车牌,效果也是不太好。前两个字还行,后面完全识别不到。这也是长文本的问题?
结果如下: ['皖', 'K'] [0.7606085 0.9683497 0.95730805] ['皖', 'K'] [0.7606051 0.9683497 0.9573083] ['啵', 'B'] [0.22663411 0.96232164 0.61064464] ['皖', 'G', '▪'] [0.4462814 0.99723685 0.8549502 0.9544695 ] ['皖', 'G', '▪'] [0.4462814 0.99723685 0.8549502 0.9544695 ] ['皖', 'A'] [0.8662637 0.84521544 0.9425101 ] ['啵', 'B'] [0.22663411 0.96232164 0.61064464] ['皖', 'A'] [0.8662637 0.84521544 0.9425101 ]
你好 问下 这个概率是怎么回事呢 两个字符为什么有三个概率呢
from attentionocr.
Related Issues (20)
- 作者训练的权重向量从docker拷贝出来上传到百度云,欢迎自取 HOT 3
- model from docker image doesnot work!
- I have some questions about training and testing
- Problems on how to calculate the probs in text recongnition
- how could i do, the train process will set on GPUS? HOT 1
- detect长度及长文本识别
- bilstm
- @zhang0jhon tensorflow run in static graph, can you give any hints on debugging during coding?
- pytorch版本实现
- There is some way to be able to execute the code without using DOCKER but only by console. I will be very grateful for your prompt response.
- steps_per_epoch根据训练集的不同需要修改吗? HOT 5
- 运行“ python train.py”时出现错误
- Network does not predict EOS token HOT 1
- $ symbol
- What's is the license information of this project?
- Formulae miscalculation
- for inception v4 stem and reduction_A the paper use valid for padding in conv3x3,why use same for padding here?
- abcnet accuracy
- 为什么不把图像裁剪下来单独训练,而是将整个图像先加载在裁剪呢?这样batch_size不就需要变小了吗? HOT 1
- train.py error ModelDescBase?
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 attentionocr.