Deep Learning - Recurrent Neural Networks with TensorFlow [Video]

Recurrent Neural Networks are a type of deep learning architecture designed to process sequential data, such as time series, text, speech, and video. RNNs have a memory mechanism, which allows them to preserve information from past inputs and use it to inform their predictions.
TensorFlow 2 is a popular open-source software library for machine learning and deep learning. It provides a high-level API for building and training machine learning models, including RNNs.
In this compact course, you will learn how to use TensorFlow 2 to build RNNs. We will study the Simple RNN (Elman unit), the GRU, and the LSTM, followed by investigating the capabilities of the different RNN units in terms of their ability to detect nonlinear relationships and long-term dependencies. We will apply RNNs to both time series forecasting and NLP. Next, we will apply LSTMs to stock “price” predictions, but in a different way compared to most other resources. It will mostly be an investigation about what not to do and how not to make the same mistakes that most blogs and courses make when predicting stocks.
By the end of this course, you will be able to build your own build RNNs with TensorFlow 2.
What You Will Learn
Learn about simple RNNs (Elman unit)
Covers GRU (gated recurrent unit)
Learn how to use LSTM (long short-term memory unit)
Learn how to preform time series forecasting
Learn how to predict stock price and stock return with LSTM
Learn how to apply RNNs to NLP
Audience
This course is designed for anyone interested in deep learning and machine learning or for anyone who wants to implement recurrent neural networks in TensorFlow 2. One must have decent Python programming skills, should know how to build a feedforward ANN in TensorFlow 2, and must have experience with data science libraries such as NumPy and Matplotlib.