lab9assignment - naveenanallamotu/Big-Data-Analytics-Lab-Assignments GitHub Wiki

Image recognition. Convolution neural networks (CNN) consist of multiple layers of receptive fields. These are small [clarification needed] neuron collections which process portions of the input image.

Data_set: Took the data from the Caltech 101 my data having the two folders it contains the train and test data. each folder has the motorcycles_sides and airplanes_side and car_markus. I took up to 100 images in the train folder for each category and 1o images in the test folder My project has two convolution and two pooling layers The pooling layer, is used to reduce the spatial dimensions, but not depth, on a convolution neural network, model, basically this is what you gain:

By having less spatial information you gain computation performance Less spatial information also means fewer parameters, so less chance to over-fit You get some translation invariance

I took only 100 in range and print the accuracy for every 50 steps and final accuracy and print the time to calculate the convent. Images are kept the documentation folder.

Cross entropy: The most common measure of error is called mean squared error. However, there are some research results that suggest using a different measure, called cross entropy error, is sometimes preferable to using mean squared error showed in the documentation folder.

Visualization and accuracy graph also reported in the tensor flow

Google conversational api:

  1. created the crossing intent

  2. Entities for Crossing has been created with different parameters and values

  3. The java rest service code has been published to heroku

Heroku is a cloud-based development platform as a service (PaaS) provider. The Heroku platform supports development in Ruby on Rails, Java, Node.js, Python, Scala and Clojure

The rest service takes the parameters and values as inputs and passes and gives the output based on those values. The below images shows the apps that are published to Heroku

While REST stands for Representational State Transfer, which is an architectural style for networked hypermedia applications, it is primarily used to build Web services that are lightweight, maintainable, and scalable. A service based on REST is called a RESTful service

  1. The Below image shows a request from api.ai is being sent to Heroku app along with the response