Description
With the exponential growth of user-generated data, there is a strong need to move beyond standard neural networks to perform tasks such as classification and prediction. Here, the go-to options are architectures such as RNNs, Gated Recurrent Units (GRUs), and Long Short-Term Memory (LSTM). Hence, for any deep learning engineer, mastering RNNs is a top priority. This course begins with the basics and will gradually equip you with the theoretical know-how and the practical skills required to build, train, and implement RNNs successfully. This course contains several exercises on topics such as gradient descents in RNNs, GRUs, LSTM, etc. This course also introduces you to implementing RNNs using TensorFlow. The course culminates in creating two exciting and realistic projects: creating an automatic book writer and a stock price prediction application. By the end of this course, you will be equipped with all the skills required to use and implement RNNs in your applications confidently.
Seat Time
11 hours 2 minutes
Languages
American English
What You Will Learn
- Utilize deep neural networks
- Define the fundamentals of RNN architectures
- Train real-world datasets using different RNN architectures
- Implement RNNs, LSTM, and GRUs through hands-on exercises
- Create and compile RNN models in TensorFlow
- Perform text classification using RNNs and TensorFlow
- Create a budget and estimates
- Work with customers and jobs
- Enter and pay bills and work with loans
- Generate reports and pay employees
Example Curriculum
- Introduction to Deep Learning Module
- Neuron and Perception
- DNN Architecture
- FeedForward FullyConnected MLP
- Calculating Number of Weights of DNN
- Number of Neurons Versus Number of Layers
- Discriminative Versus Generative Learning
- Universal Approximation Theorem
- Why Depth
- Decision Boundary in DNN
- Bias Term
- Activation Function
- DNN Training Parameters
- Gradient Descent
- Backpropagation
- Training DNN Animation
- Weight Initialization
- Batch MiniBatch Stocastic
- Batch Normalization
- Rprop Momentum
- Convergence Animation
- DropOut EarlyStopping Hyperparameters