The Progress of Products Classification

Cause now we are considering to classify the product by two kinds of features, product images, and product title. I tried to handle these two kinds of features individually, on the product title side, I used Keras build a simple RNN model for classifying 10 classes product, and I got a good result, about 98% accuracy. I test the model with some products from our site, except the title is too ambiguous I can get a proper result, the model doesn't know how to handle some combined word, e.g. 'SmartWatch'. But I found that the product images are very clear, so I wonder if I could combine these two features it wouldn't be a big problem. you can see the watch at  , and my model recognized it as a motherboard. 

On the other side, I want to build a model to classify the product images. Different from usual image classification problem, I'm going to make a classifier working on a set of images, for example, a Lenovo Laptop product would contain an image of Lenovo logo, the laptop's front and back photograph, and all images can in any order. So, I'm just doing a job with a set of non-sequential data.

Three failed attempts

1.Working on a single image and combine the result

I trained a usual classifier that accepts a single image, I wrote the model with Keras Vgg16 like before. Suppose we have 3 images, I pass each image to the model, and I got a probability distribution of all classes, assume we have 4 classes, for each image I would get a probability vector like [0.1,0.8,0.05,0.05]. Then, I use weighted average to merge all probability, and I got a problem, If I have 3 images one image is ambiguous and get a low rank on the right classes, suppose the first class is the right class[0.1,0.4,0.3,0.3], and the other two images I get a high rank in the first class [0.98,0.0001,0.003,0.016], for a human, it's very certain this product belongs to the first class, but after weighted average the probability might like[0.68,0.1,0.05,0.03].

I also try to build a simple RNN model which accepts all probability vectors, and it didn't work.

2.Combine all images into a single data block

Most product images are RGB image, from a mathematic view, it's a 3rd order tensor with shape (3,width,height), and each element in the tensor is an integer from 0 to 255.

First, I convert all images into a grayscale image, now the image's shape is (width, height), it's a matrix. I limit a max number of images as N, if the number of images is less than N, I would fill some blank images, a matrix with all elements set to zero. Second, I merge these images on the 3rd axis, after that, I got a tensor with shape (N, width, height), Finally, I build a model can accept the tensor. But I failed, I got a different result when I reorder the images.

I think the reason why I failed is after convolution and pooling layers I get a 3rd order tensor, I need to reshape the tensor to a vector and pass it to the final classifier, that's the job the Keras Flatten layer did, and it's more like a weighted average job. when I change the order of the images, I would get a different vector before the classifier.

3.Add attention mechanism to the model

As I mentioned above, the weighted average caused the problem, I want to do something prevent weighted average before Flatten layer. Attention mechanism is a new technique always be used in RNN, it can make the model learn which part is more important and pay attention to that part. I flowed keras-attention-mechanism to add the attention mechanism to my model. But I failed like before.

Attention mechanism can't promise to pass a same tensor to the classifier with a different order of images.

Some thoughts

Like this paper mentioned, I think to deal with non-sequential data, we need to use some statistics feature.

 

Three failed attempts of handling non-sequential data的更多相关文章

  1. Time Series data 与 sequential data 的区别

    It is important to note the distinction between time series and sequential data. In both cases, the ...

  2. Open-sourcing LogDevice, a distributed data store for sequential data

    https://logdevice.io/blog/2018/09/12/open-sourcing-announcement.html September 12, 2018   We are exc ...

  3. ElasticsearchException: java.io.IOException: failed to read [id:0, file:/data/elasticsearch/nodes/0/_state/global-0.st]

    from : https://www.cnblogs.com/hixiaowei/p/11213143.html 1.以前装过elasticsearch,重新安装elastic search ,报错 ...

  4. PRML读书会第十三章 Sequential Data(Hidden Markov Models,HMM)

    主讲人 张巍 (新浪微博: @张巍_ISCAS) 软件所-张巍<zh3f@qq.com> 19:01:27 我们开始吧,十三章是关于序列数据,现实中很多数据是有前后关系的,例如语音或者DN ...

  5. The Swiss Army Knife of Data Structures … in C#

    "I worked up a full implementation as well but I decided that it was too complicated to post in ...

  6. LOAD DATA INFILE Syntax--官方

    LOAD DATA [LOW_PRIORITY | CONCURRENT] [LOCAL] INFILE 'file_name' [REPLACE | IGNORE] INTO TABLE tbl_n ...

  7. redis.clients.jedis.exceptions.JedisConnectionException: java.net.SocketException: 断开的管道 (Write failed)

    昨晚,包发到测试环境中,出现redis.clients.jedis.exceptions.JedisConnectionException: java.net.SocketException: 断开的 ...

  8. troubleshooting-执行Oozie调度Hive导数脚本抛java.io.IOException: output.properties data exceeds its limit [2048]

    执行Oozie调度Hive导数脚本抛java.io.IOException: output.properties data exceeds its limit [2048] 原因分析 shell脚本中 ...

  9. Analyzing Microarray Data with R

    1) 熟悉CEL file 从 NCBI GEO (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE24460)下载GSE24460. 将得到 ...

随机推荐

  1. mybatis 类创建流程

    Configuration ---> XmlConfigBuilder ---> SqlSessionFactoryBuilder ---> SqlSessionFactory(Co ...

  2. 2.ReactJS基础(虚拟DOM,JSX语法)

    将脚手架(create-react-app)创建的todolist项目精简为hello world示例 即,删除自动生成的样式文件.logo.svt.App.test.js.serviceWorker ...

  3. 数据分析处理库--Pandas

    Pandas库: pandas索引与计算:

  4. java-16习题

    编写程序,产生10组彩票的“35选7”玩法的7个随机数.(-)随机数不能重复. 范围[,) import java.util.Iterator; import java.util.Random; im ...

  5. python中的Matplot库和Gdal库绘制富士山三维地形图-参考了虾神的喜马拉雅山

    首先请大家读一下面这篇文章了解什么是Gdal http://blog.csdn.net/grllery/article/details/77822595 剩下的我要公布绘制富士山的代码了,虽然基本co ...

  6. 服务器、应用框架、MVC、MTV

    web服务器:负责处理http请求,响应静态文件,常见的有Apache,Nginx以及微软的IIS. 应用服务器:负责处理逻辑的服务器.比如php.python的代码,是不能直接通过nginx这种we ...

  7. 学习笔记TF054:TFLearn、Keras

    元框架(metaframework). TFLearn.模块化深度学习框架,更高级API,快速实验,完全透明兼容. TFLearn实现AlexNet.https://github.com/tflear ...

  8. 第五章Bookstrap

    响应式原理: @media screen and (min-width:300px) and (max-width:500px) { /* CSS 代码 */ } #代表页面宽度大于300px和小雨5 ...

  9. mybatis 中javaType和OfType 的区别

    JavaType和ofType都是用来指定对象类型的,但是JavaType是用来指定pojo中属性的类型,而ofType指定的是映射到list集合属性中pojo的类型.pojo类: publiccla ...

  10. [工作积累] Tricks with UE4 PerInstanceRandom

    最近在用UE4的Instancing, 发现限制很多. Unity有instancing的attribute array (uniform/constant buffer),通过InstanceID来 ...