Chevron Left
返回到 Anomaly Detection in Time Series Data with Keras

學生對 Coursera Project Network 提供的 Anomaly Detection in Time Series Data with Keras 的評價和反饋

120 個評分
32 條評論


In this hands-on introduction to anomaly detection in time series data with Keras, you and I will build an anomaly detection model using deep learning. Specifically, we will be designing and training an LSTM autoencoder using the Keras API with Tensorflow 2 as the backend to detect anomalies (sudden price changes) in the S&P 500 index. We will also create interactive charts and plots using Plotly Python and Seaborn for data visualization and display our results in Jupyter notebooks. 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 Keras pre-installed. 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....



Jun 02, 2020

This guide is incredible, easy to understand for beginners in the field like me, I'm really grateful because it helps me a lot.


Jun 09, 2020

It is good. step by step so I can understand. but unfortunately there are no subtitle.


1 - Anomaly Detection in Time Series Data with Keras 的 25 個評論(共 31 個)

創建者 Joerg A

May 24, 2020

The quiz has a question about "image anomaly detection" - not fitting the course.

There is little explanation, especially what the influence of the parameters are. Most of the time one spends typing and the aloted rhyme time is way too short. I was watching at 2x speed (an annotation told to use 1.5x speed) AND WAS STILL NOT FINISHED when the allotted time was over.

Think hard if you want to pay money for that stressed learning pace.

創建者 Amit D

May 31, 2020

Pathetic experience! I see that this is a good idea executed poorly. Clearly Rhyme is not ready yet. I experienced the following issues. 1. Long connection times for Rhyme. In many cases, the desktop didn't even connect. 2. The keyboard of your machine is not synced with Rhyme's desktop. So, certain keys (e.g. CAPS LOCK) don't work. 3. The instructor doesn't explain why he is doing something, just does it.

Coursera should apologize for wasting my time.

創建者 Wim P

May 30, 2020

The above rating has nothing to do with the actual content, but has everything to do with the flawed system used to bring the content.

After trying to start this one hour course for over three hours, I gave up. The product (Rhyme) is clearly not ready, so it's not worth your money (yet?).

If you compare the value you get out of these courses versus the other courses on the platform, this should actually be zero out of five.

創建者 swarnima

May 06, 2020

It is one of the best guided project I came through on coursera. The project is of intermediate level, quite clear and understandable. The instructor from rhyme was quite good. He explained every part, every function and reason behind their use quite clearly. I recommend this to anyone who is already into time series forecasting and wants to improve his/her skills into it.

創建者 Octavio A T N

Jun 02, 2020

This guide is incredible, easy to understand for beginners in the field like me, I'm really grateful because it helps me a lot.

創建者 Suci K P

Jun 09, 2020

It is good. step by step so I can understand. but unfortunately there are no subtitle.

創建者 Jacoby W

May 19, 2020

I love how well he explained everything and made it simple to follow

創建者 Adonia S

Jun 13, 2020

Excellent way of teaching and also good teacher as well

創建者 Deepak G

Jun 13, 2020

Very well explained. Instructor knows what to do.

創建者 khushbu G

May 30, 2020

An informative course, worth spending time on!

創建者 Gangone R

Jul 03, 2020

very useful course

創建者 Rishabh R

May 06, 2020

Excellent project

創建者 Dr. H M

May 22, 2020

thank u coursera


Jun 02, 2020

Very Helpful !

創建者 Reinhold L

May 02, 2020

Useful example

創建者 Josafat E T

Jul 31, 2020

The greatest

創建者 k j

Jul 01, 2020


創建者 sarithanakkala

Jun 25, 2020


創建者 p s

Jun 21, 2020


創建者 Rifat R

Jun 07, 2020


創建者 Ashwin P

May 21, 2020


創建者 George X H

Jun 23, 2020

A little bit of more explanations on the autoencoders on what each components and each line of code does will help. Also a little bit of summary on what the results means for S&P data would be better too. For example, anomalies that we detected does not just mean sudden jumps in S&P closing price levels, it means the changes that are not predicted by our neural network. So if there's a big jump on index prices, if it's predicted by our RNN it wouldn't count as an anomaly.

創建者 Vikram I

Jul 28, 2020

Rhyme is Bogus. The instructor did a fairly good job. Good for a quick refresher. Do not expect any conceptual stuff. It is for people who know the theory and need some practice doing applied DL on Time Series.

創建者 Amitesh S

Jul 24, 2020

The project was useful. Rhyme's interface needs testing and an upgrade.

創建者 Md A R

Jun 09, 2020

Good.. Need more explanation of code...