GithubHelp home page GithubHelp logo

Comments (8)

yalecyu avatar yalecyu commented on May 26, 2024

caffe环境上和warpctc-caffe没有区别,网络框架是 cnn-lstm-ctc,warpctc-caffe是 lstm-ctc,所以 tree/master/src 是没有区别的

from crnn.caffe.

piaobuliao avatar piaobuliao commented on May 26, 2024

哦,谢谢,好像warpctc-caffe的输入LSTM的就直接是原图哦,那你这个网络结构和
https://github.com/dlunion/CaffeLSTM-OCR 的识别验证码的差不多,不过我现在在用https://github.com/dlunion/CaffeLSTM-OCR 做,它那个项目输入LSTM的没有像你一样有个Reshape层,但是也是能收敛的哦,你具体可以看下 dlunion/CaffeLSTM-OCR#1 我的问题哦

from crnn.caffe.

yalecyu avatar yalecyu commented on May 26, 2024

Reshape层只是reshape对网络效果没有影响。这个是cnn处理图像得到特征向量输入到LSTM

from crnn.caffe.

piaobuliao avatar piaobuliao commented on May 26, 2024

你这个输入LSTM的Shape是 32x64x512 (time_step:32 batch_size:64)
CaffeLSTM-OCR 的输入LSTM的Shape 19 1 512 8 (time_step:19 batch_size:1)

https://zhuanlan.zhihu.com/p/28054589 按照这里介绍RNN结构的,你的这个输入xi是1x512的,而CaffeLSTM-OCR 的输入是1x512x8的? 还是就是1*512的(后面8的信息没了)? 不知道LSTM的输入到底是需要什么Shape哦。

from crnn.caffe.

piaobuliao avatar piaobuliao commented on May 26, 2024

那输入LSTM的只是是向量? 2D的矩阵不行的吗? 还有你做车牌识别的话 time_step是设置了为32喽?我现在在搞银行卡号识别,time_step设置八九十收敛很慢,设置四五十好像识别率又不是很高

from crnn.caffe.

yalecyu avatar yalecyu commented on May 26, 2024

因为某个维度为1,就可以把这个维度删掉了,我的reshape做得是这个工作。cnn的输出是64x512x1x32(Batch_sizie x num_output x h x w)

from crnn.caffe.

piaobuliao avatar piaobuliao commented on May 26, 2024

那假如cnn的输出是64x512x4x32(Batch_sizie x num_output x h x w)
,那经过permuted_data层输入LSTM的是32x64x512x4,这样也是可以的吗?

from crnn.caffe.

yalecyu avatar yalecyu commented on May 26, 2024

我觉得可以,你可以试试。

from crnn.caffe.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo 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.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.