DEEP LEARNING

RNN and CNN

The repository contains two projects - 1)Based on CNN(Counvolutional neural network)

                                 2)Based on RNN(Recurrent neural network)

1) CNN :

This model identifies flowers in an image!!

Model classifies feeded image into 5 classes - > 1)Daisy > 2)Dandelion > 3)Sunflower > 4)Rose > 5)Tulip

 Directory contains -

Code which downloads data from Kaggle , segregates data into training and validation set , preprocessing image, building model, training , testing with manually collected data from another source, plotting graph, and lets you to upload and test model

A ppt ( explaining structure of neural network , detailed info , graphs , images)

2) RNN :

A model which can guess your emotions from sentence!!!

Model basically reads the sentence and predicts responce- POSITIVE/NEGATIVE/NEUTRAL

 This directory contains-

Code which loads dataset , preprocessing(tokenizer,embedding) , model building, training, testing, graph plotting(accuracy and loss)

A ppt (which explains detailed version of project , graphs, gif)

Datatset

Dataset is collected from various sources - kaggle , imdb , tweets