Chevron Left
返回到 Explainable AI: Scene Classification and GradCam Visualization

學生對 Coursera Project Network 提供的 Explainable AI: Scene Classification and GradCam Visualization 的評價和反饋

4.7
46 個評分
8 條評論

課程概述

In this 2 hour long hands-on project, we will train a deep learning model to predict the type of scenery in images. In addition, we are going to use a technique known as Grad-Cam to help explain how AI models think. This project could be practically used for detecting the type of scenery from the satellite images....

熱門審閱

篩選依據:

1 - Explainable AI: Scene Classification and GradCam Visualization 的 8 個評論(共 8 個)

創建者 Vipul G

2020年7月27日

I like the course, it is exceptional.

But if you provide the materials(train/test files) to download it will be better to apply it on our own

創建者 Alexandros O

2020年12月25日

(+) Very insightful introductory project course to CNN and XAI. The instructor was explaining as much as possible to all parts. Providing such images was really helpful.

(-) There were several mistakes in the code. A prerequisite for this course could also be the mathematical background and thus, more explanation on why and how each mentioned-part could be provided. Not all explanation parts for XAI are provided to jpnb for students.

創建者 Yaron K

2021年9月26日

A step by step explanation of how to build a Resnet Image Classification Convolutional Neural Network. Including how to use a technique known as Grad-Cam to visualize how different parts of the image effect the final classification.

Cons: No theory. It shows all the pieces of a working model. But not WHY it works.

Note: the notebook in Files is empty. The mostly complete notebook is in Files-->Notebooks

創建者 Jesus M Z F

2020年7月19日

Excelente curso, Muchas gracias

創建者 Stud 2

2020年8月1日

very helpful

創建者 Kamlesh C

2020年7月27日

thanks

創建者 Samy S S E

2020年8月26日

it's an exciting course it covers all machine learning life cycle steps in a short time and organizable way

創建者 Simon S R

2020年9月2日

This project should be more about GradCam Visualization and should dive deeper into its details, but not provide an explicit overview of all the steps necessary to build the original model.