Chevron Left
Back to Simple Recurrent Neural Network with Keras

Learner Reviews & Feedback for Simple Recurrent Neural Network with Keras by Coursera Project Network

4.4
stars
120 ratings

About the Course

In this hands-on project, you will use Keras with TensorFlow as its backend to create a recurrent neural network model and train it to learn to perform addition of simple equations given in string format. You will learn to create synthetic data for this problem as well. By the end of this 2-hour long project, you will have created, trained, and evaluated a sequence to sequence RNN model in Keras. Computers are already pretty good at math, so this may seem like a trivial problem, but it’s not! We will give the model string data rather than numeric data to work with. This means that the model needs to infer the meaning of various characters from a sequence of text input and then learn addition from the given data. This course runs on Coursera's hands-on project platform called Rhyme. On Rhyme, you do projects in a hands-on manner in your browser. You will get instant access to pre-configured cloud desktops containing all of the software and data you need for the project. Everything is already set up directly in your internet browser so you can just focus on learning. For this project, you’ll get instant access to a cloud desktop with Python, Jupyter, and Tensorflow pre-installed. Please note that you will need some experience in Python programming, and a theoretical understanding of Neural Networks to be able to finish this project successfully. Notes: - You will be able to access the cloud desktop 5 times. However, you will be able to access instructions videos as many times as you want. - This course works best for learners who are based in the North America region. We’re currently working on providing the same experience in other regions....

Top reviews

Filter by:

1 - 17 of 17 Reviews for Simple Recurrent Neural Network with Keras

By Pravin S

•

May 21, 2020

Best Understanding of Recurrent Neural Network in simplest way.

By SENTHIL K B

•

May 12, 2020

Excellent planning and guidance throughout

By Gangone R

•

Jul 4, 2020

very useful course

By Prakash S

•

May 31, 2020

Excellent tool

By Kamlesh C

•

Jun 21, 2020

Thanks

By Abel F Z C

•

Jul 9, 2020

good

By Vajinepalli s s

•

Jun 20, 2020

nice

By Ashwin P

•

May 12, 2020

good

By Daniel S R

•

Jul 13, 2020

Good guided course. I would add a quite more deep details in the model architecture to understand better how are the inputs and the outputs of each layer in the RNN model

By Mohammed B

•

Jun 24, 2020

good

By Mónika J

•

May 13, 2020

I think that the explanation on the code is not enough for beginners and that it mostly depends on the student's background and effort wether they understand it or not.

By Salil M

•

May 22, 2020

The knowledge about RNN was average, it was mainly focusing on data processing for RNN use, can be improved by using RNN more rigorously

By Beemen B

•

Nov 7, 2022

I am rewriting the review , as I have changed my mind about the project.

Initially, the project seemed a bit too simple/naive.

After finishing it, however, I was able to dig a lot deeper into online articles, and gain a much better understanding of RNN's.

In brief, it was a great door opener to the world of RNN's

By M V

•

Sep 25, 2020

Awesome course, really learnt a lot !

By Pramod H K

•

Jul 26, 2020

Very good and simple intro to RNN.

By Dr R S

•

May 10, 2020

Will learn the PYTHON soon and get expert in this.

By Sidney V

•

Mar 24, 2024

Positive point: 1) The application example, despite being simple, is interesting from a didactic perspective. Negative points: 1) The processing logic of the RNN was explained vaguely. For example, it was not well explained why the model used must be formed by a combination of "encoder" & "decoder". It was also not clearly explained how we can check the number of inputs and outputs of each layer of the sequential model, to facilitate understanding of the model. 2) No figure was shown to illustrate the neural network. This is a serious deficiency in a quick course like this, as the illustration of the neural network would make understanding the model much easier. "A picture is worth a thousand words". 3) The cloud platform was very slow, making it almost unfeasible to run the notebook. Because of this, I had to download the notebooks and run them on my own computer, with additional work to install all the necessary Python modules that I wouldn't need to install if I worked directly on the cloud platform. 4) The Python codes in the notebook were created 5 years ago (according to information in the Jupyter "Files" tab, which said "Last Modified: 5 years ago"). Keras has had many updates since then, and the codes used in course are already obsolete. More specifically, the model creation code (Task 3) and the training code (Task 6) did not work with the current version of Keras on my computer. I was able to complete the course and get the certificate just because I did well in the Quiz, but I couldn't implement the model using the outdated codes. I've already spent a lot of time trying to rewrite the obsolete parts of the code. Why the instructor doesn't keep the codes updated? At the very least, there should be a new version of the notebook for download with updated codes. 5) I posted a request for help on the Forum. After 10 days, I have not received any reply. What caught my attention was the fact that the course has 6,085 "enrolled students", but the most recent posts on the course forum are from some years ago. How is it possible that a course with so many students supposedly enrolled has an inactive forum? 6) Sorry, but the point that I consider most negative is having to pay for a course with all the negative points mentioned above. I'm sure that if all the negative points are solved, this course will be a good option to teach RNN.