Each project comes with 2-5 hours of micro-videos explaining the solution.
Code & Dataset
Get access to 50+ solved projects with iPython notebooks and datasets.
Add project experience to your Linkedin/Github profiles.
Understanding the problem statement
Importing the problem statement
Installing Keras and LSTM
Importing the necessary libraries for applying Neural Networks
What are Recurrent Neural Networks and how do they work
Understanding basics of NLP
Performing basic EDA and checking for the null values
Making your own Neural Network from scratch
Applying LSTM without dropout and evaluating the result
Applying LSTM with dropout and evaluating the result
Creating the model with double drop out, drop out between layers and drop out within layers of LSTM
Introducing the concept of the Fully connected network to optimize the model further
Finally evaluating the model
Making predictions for the test Dataset
A sequence to sequence prediction for developing a classification system is of very much required in developing applications. Standard approaches for developing applications won't help in providing accuracy. Hence, as an example let's take an IMDB movie review dataset and create some benchmarks by using RNN, RNN with LSTM and drop out rate, RNN with CNN, and RNN with CNN plus drop out rate to make a composite sequence to sequence classification work. We can compare the model accuracy as well.