Künstliche Intelligenz und Maschinelles Lernen in der Praxis

1. 00:00Let's look now I'm going to just close.
2. 00:02what we did in week 4.
3. 00:06We have developed a Project developed an AI,
4. 00:12sign language on images in the alphabet from A to Z. And they had
5. 00:19own models and from the ground up.
6. 00:24But also some pre-trained AI models used, i.e. transfer learning applied
7. 00:29and saw that this is a neat performance increase.
8. 00:35We also have the possibility of We looked at data gathering and saw it.
9. 00:39for example, the possibility of Data acquisition gives you that you can get historical data
10. 00:44that you can use web crawling , for example from the Internet
11. 00:50Collect information programmatically.
12. 00:53Or that you can, for example, create data streams. as an API.
13. 01:00The learning objectives for this week were once the Introduction to Computer Vision,
14. 01:04So, image processing. We have the Use case in image recognition also with pretained
15. 01:10Machine learning models.
16. 01:12That means we wanted to give you an understanding to teach how such a thing can look,
17. 01:17You don't always have to train from scratch. but can also use transfer learning.
18. 01:22And also simply because there is such a is an important issue and should not be neglected,
19. 01:27the issue of data collection, i.e. What a data acquisition might look like?
20. 01:32and how important it is that you're also concerned about that.
21. 01:36Here too, at this point again the reference to deep learning Computer Vision course for openHPI students
22. 01:45If you are interested in the subject, again our recommendation,
23. 01:49Please feel free to drop by.
24. 01:52That's what it was with week four.
25. 01:54We would like to join a few address final words to you
26. 01:57and thank you for taking this course.

To enable the transcript, please select a language in the video player settings menu.