Sequences, Time Collection and Prediction

About this Course

In case you are a software program developer who needs to construct scalable AI-powered algorithms, you should perceive use the instruments to construct them. This Specialization will educate you finest practices for utilizing TensorFlow, a well-liked open-source framework for machine studying.

On this fourth course, you’ll learn to construct time sequence fashions in TensorFlow. You’ll first implement finest practices to organize time sequence information. You’ll additionally discover how RNNs and 1D ConvNets can be utilized for prediction. Lastly, you’ll apply every thing you’ve realized all through the Specialization to construct a sunspot prediction mannequin utilizing real-world information! The Machine Studying course and Deep Studying Specialization from Andrew Ng educate a very powerful and foundational ideas of Machine Studying and Deep Studying. This new deeplearning.ai TensorFlow Specialization teaches you use TensorFlow to implement these ideas so to begin constructing and making use of scalable fashions to real-world issues. To develop a deeper understanding of how neural networks work, we suggest that you simply take the Deep Studying Specialization.

WHAT YOU WILL LEARN

• Remedy time sequence and forecasting issues in TensorFlow

• Put together information for time sequence studying utilizing finest practices

• Discover how RNNs and ConvNets can be utilized for predictions

• Construct a sunspot prediction mannequin utilizing real-world information

SKILLS YOU WILL GAIN

• Forecasting
• Machine Studying
• Tensorflow
• Time Collection
• prediction

Syllabus – What you’ll be taught from this course

Content material Ranking97%(9,632 rankings)

WEEK

1

6 hours to finish

Sequences and Prediction

Hello Learners and welcome to this course on sequences and prediction! On this course we’ll check out a few of the distinctive concerns concerned when dealing with sequential time sequence information — the place values change over time, just like the temperature on a specific day, or the variety of guests to your site. We’ll talk about varied methodologies for predicting future values in these time sequence, constructing on what you’ve realized in earlier programs!

WEEK

2

6 hours to finish

Deep Neural Networks for Time Collection

Having explored time sequence and a few of the frequent attributes of time sequence corresponding to development and seasonality, after which having used statistical strategies for projection, let’s now start to show neural networks to acknowledge and predict on time sequence!

WEEK

3

5 hours to finish

Recurrent Neural Networks for Time Collection

Recurrent Neural networks and Lengthy Quick Time period Reminiscence networks are actually helpful to categorise and predict on sequential information. This week we’ll discover utilizing them with time sequence…

WEEK

4

7 hours to finish

Actual-world time sequence information

On prime of DNNs and RNNs, let’s additionally add convolutions, after which put all of it collectively utilizing a real-world information sequence — one which measures sunspot exercise over a whole lot of years, and see if we will predict utilizing it.