Okay, so hello everyone. Just quickly let's go a little bit above the evaluation. So the

evaluation was actually quite fine. I mean we had, interestingly, the most people come

from informatics. That's nice. So I mean in the first side this was only an informatics course

and then many more people asked or actually are asked if other study courses also can use this

class in their study program. So you still see that it's quite diverse. So from the 31 who

participated about a third is informatics and the other ones are totally different study courses

here. The most are from last semester's or master's which is also fine I think. I wonder

why. Okay, apparently there are several more people from the master's actually than from

the bachelor's. That's a little bit strange because it's actually more meant to be a bachelor's

course but I mean I'm totally fine for that. I don't care. Yeah, I think the rest was not so

interesting. Yeah, this is maybe interesting. How many have participated in presence and here it

seems the most who also evaluated the course also were somewhat in presence and then we had

several so yeah, who did? Asynchronous? No, this is synchronous. That doesn't work but asynchronous

we also had several participants here. I don't know. Yeah, actually this question doesn't make

sense because there is no synchronous online. No, well in the exercise. Okay, then we had some

about the organization and so on. I think that's rather fine. We have some outliers here which I

hope to improve of course and maybe which is a little bit more challenging or I think where I

have to work really is just on this point. So that to make it the course more, yeah,

I don't know the word in English, so that you basically while teaching that you can follow

the course and or the lectures in a better way. So here I think there is much potential for me to

improve this course and okay we always have some outliers but this is I think this is in every

lecture the case. So the most people were actually rather happy with this course so thank you for

that and yeah I guess also in this part and this says about the difficulty we are still here above

them. I mean the best would be three now but we are still more leaning towards that it's too

difficult. That's of course always challenging. I mean especially with so many different courses,

study programs, yeah it's for some people it is too difficult, too much math for them, for some

it's too less math so yeah it's a little bit challenging but I see that overall I think I can

still improve here to make it a little bit less math heavy but it's actually also mainly in the

later parts I would say and in the beginning. Yeah okay I guess I can improve that and yeah then we

had some some comments here so thank you for that. So the most people liked these quizzes so there

were some comments about these quizzes so I think I will I will read you know do that further in the

next iterations and also the people like that it's that it's recorded so yeah I think both parts can

be can be also done in the next year so there's now one semester of a break where intro ML is not

taking place and then we have some things to improve and these are always very interesting

yeah so somebody says in respect to the exercises so I think there the problem was that you only

that we currently had two deadlines kinda or and instead it would be better to only have a deadline

kind of for the for the exercise submission and then not a deadline for the you know when you talk

about the exercise I think that's reasonable I think we can change that makes sense to me okay

then to mark what is interesting and what is net not interested which is just a derivation yeah I

don't know how I can mark that but yeah I'm not so sure how to practically do that but I typically

say when I think it's into important typically derivations the most derivations you don't need

to learn now for the exam so okay some are complaining about too much pre image processing

so no okay it's more ironically probably meant yeah so two points here in this so so yeah there

is some this there is a lot pre-proced image pre-processing in my goal is actually to to

shrink the image processing part further because I also have the feeling on the other side it's

very good if you know this image processing if you are then at our lab if you are a master thesis

or bachelor thesis typically they are somewhat related to image images yeah and so I don't want

to kick it totally out so make it totally classifier based because for that we have the

lecture pattern recognition yeah so that that doesn't make much sense so I think I will still

keep it but maybe remove some I already removed some parts here but there could maybe be dropped