Skip to content

This repository contains all the projects carried out with respects to learning and experimenting on Deep Learning and Neural Networks. Deep Learning models are created regarding predictions based on convolutional and recurrent neural networks.

License

Notifications You must be signed in to change notification settings

gauravsharma884/Deep-Learning

Repository files navigation

Deep-Learning-And-Natural-Language-Processing

This Repository contains the following files

  1. Convolutional Neural Network.ipynb
  2. Celebrity_Recogniton.ipynb
  3. Poetry_Generator.ipynb

Convolutional Neural Network (Convolutional Neural Network.ipynb)

This project is carried out for undertanding convolutional neural network.

About Project:

In this script Convolutional Neural Network Model is build to classify and recognize images of cat and dog.

Marvel Star-Cast: Celebrity Recognition (Celebrity_Recogniton.ipynb)

This project is performed to recognize and predict celebrities from the Marvel Star-Cast

About Project

In this script a model is build to recognize celebrity faces The Marvel Starcast using CNN (Convolutional Neural Networks). For this Project 4 celebrities considered for building the model

  1. Chris Evans: Captain America
  2. Chris Hemsworth: Thor
  3. Paul Rudd: Ant-Man
  4. Robert Downey Jr.: Iron-Man

Poetry Generation (RNN + Markov) (Poetry_Generator.ipynb)

This Projects builds model to generate new verses in the style of the poetry by using RNN (Recurrent Neural Networks) and Markov Chains.

About Project

In this script a model is build to genrate a new poetry lyrics, the vocabulary and rhyme schemes of rappers: Notorious B.I.G. and Lil Wayne are taken into consideration

The Keras.LSTM functions (Recurrent Neural Networks) are used to predict the properties of the next line of the poem (for example, the number of syllables and the rhyme scheme), so that an appropriate new sentence can be chosen. The Markovify functions (Markov Chains) are used to build new sentences (based off of word1=>word2 probabilities) from the setences that were generated with Markovify.

About

This repository contains all the projects carried out with respects to learning and experimenting on Deep Learning and Neural Networks. Deep Learning models are created regarding predictions based on convolutional and recurrent neural networks.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published