[Tensorflow2] Complete conquest of Python deep learning - Latest techniques of GAN, BERT, RNN, CNN

Deep learning is no longer an option, it’s a necessity.
Are you going to learn rapidly developing deep learning only with old textbooks and techniques?

[Tensorflow2] Complete conquest of Python deep learning
Latest techniques of GAN, BERT, RNN, CNN

🗒 Course Introduction

As AlphaGo conquered the human domain of Go, we learned that the future of artificial intelligence has become a reality.

Meanwhile, artificial intelligence, machine learning, and deep learning are receiving tremendous attention and developing rapidly.

However, there are many cases where people do not even know how to use deep learning in practice.

Despite the rapid development of technology, I see that there are only textbooks that teach CNN and RNN techniques, which are already traditional techniques, despite being a fast-paced field.

We have created a course that teaches all of the latest deep learning techniques: GAN, BERT, RNN, and CNN.

We created a course that fosters practical skills by working on various project examples as well as theories.

🌈 Project Introduction

Project 1. Using CNN (Convolutional Neural Network) technique

Create a deep learning model to recognize MNIST handwritten digit images.

Project 2. Using the RNN (Recurrent Neural Network) technique

Create a deep learning model that analyzes IMDB movie review data.

Project 3. Using the new technology of natural language processing, BERT ( Bidirectional Encoder Representations from Transformers )

We build a deep learning model that analyzes IMDB movie review data with 94% accuracy.

Project 4. Using the latest GAN (Generative Adversarial Networks) technique

Create a deep learning model to generate MNIST handwriting images.

Special lecture

'Model accuracy is over 99%  We have added a special lecture called 'Raising the bar'. This lecture is ' [ Raspberry Pi ] IoT Deep Learning Computer Vision Practice  The project started with a question from students in the course: "Why can't the MNIST handwriting model say '7' is '7'?" Of course, there are various factors such as the accuracy of the model, program exception handling, and the raw MNIST data, but since the existing Nueral Network model is simple for learning, I created the content to reconfigure the Nueral Network model to increase its accuracy to 99.38%.

🙋🏻‍♂️ Questions about the lecture

Q. What program does this course use?
A. This course uses Python 3, Anaconda, and TensorFlow 2 as its basics.

If you have taken the prerequisite course, Python Machine Learning, you are using the same program.

Q. What are the features of this course?
A. Deep learning is developing rapidly, so you need to learn the latest techniques. In this course, you will learn the latest techniques such as GAN and BERT.

This course will teach you deep learning through practical projects as well as theoretical explanations.

Q. Can non-majors also take the course?
A. Deep learning and data science are not fields that only those with a computer science major can pursue.

This is something you can learn and use if you have the passion.