|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ENKÄTER | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Visa resultatHär kan se resultatet från utvärderingen och exportera statistiken till ett annat program. Det går också att göra en enkel filtrering genom att klicka på svarsalternativen och kommentarerna eller en avancerad filtrering genom att använda knappen längst ned.
LP1 H12 FFR135 Artificial neural networksStatus: AvslutadÖppen för svar: 2012-10-26 - 2012-11-11 Antal svar: 35 Procent av deltagarna som svarat: 55% Kontaktperson: Linn Warg»
Your own effort1. How many hours per week did you spend on this course?We mean total time, that is, it comprises the time you spent in class and the time you spent on your own work. Try to estimate the average time over the entire study period.35 svarande
Genomsnitt: 3.62 - wasn"t getting any where» (At most 15 hours/week) 2. How large part of the teaching offered did you attend?35 svarande
Genomsnitt: 4 - I missed a few lectures due to teaching obligations.» (75%) Goals and goal fulfilmentThe course syllabus states the course goals in terms of learning outcomes, i.e., knowledge, skills and attitudes to be acquired by the student during the course.3. How understandable are the course goals?35 svarande
Genomsnitt: 2.57 - There should be a better information for people not going on the lectures. Some of the info given at the lectures was not only teaching but information on how to do the example sheets which in my opinion should also be posted on the webpage. For example, tips on how to plot the results, minimum results to show etc etc.» (The goals give some guidance, but could be clearer) 4. Are the goals reasonable considering your background and the number of credits?Answer this this question and the succeeding one, only if you do know the course goals.32 svarande
Genomsnitt: 2.18 - Why can I not untick this question? As it should be left blank if I have not read the goals...» (Yes, the goals seem reasonable) 5. Did the examination assess whether you have reached the goals?31 svarande
Genomsnitt: 2.74 - Not enough feedback on the work, so basically didn"t know if you got it right or not.» (To some extent) Teaching and course administration6. To what extent has the teaching been of help for your learning?35 svarande
Genomsnitt: 2.57 - I tended to get lost in huge chunks of equations. I believe some of the proofs could be left as a homework, and the spared time could be used on more practical tasks.» (Some extent) 7. To what extent has the course literature and other material been of help for your learning?35 svarande
Genomsnitt: 2.34 - I"ve barely used the book from Haykin and much of the material is from a book which had to be bought from Amazon, which is tiresome. Rework the handouts so it references more to Haykin, for additional reading. The notes are very thin.» (Small extent) 8. How well did the course administration, web page, handouts etc work?35 svarande
Genomsnitt: 3.28 - The points needed for higher grades were not displayed on the web page» (Rather well) Study climate9. How were the opportunities for asking questions and getting help?35 svarande
Genomsnitt: 3.34 - The only time when there was an opportunity to ask questions was in lectures. But when 50% of the class had questions after lecture, there was really no room for discussion and you felt that you have to stress an answer.
There was also office hours every friday. Since they were located the day after the hand in of the previous examples sheet, you didn"t really have had the time to dig deep into the exercise and run into any bigger problems which generates questions.
Then one can argue that you should start earlier with the next examples sheet, but really there wasn"t time for that. At least I did not have time for that.
I believe the course needs a lot more time for discussion. The examples classes helped to understand what should be in your report, but it did not help understand the problem very much.» (Very poor) 10. How well has cooperation between you and your fellow students worked?35 svarande
Genomsnitt: 3.54 - I don"t know anyone in the course and thus had very little chance to.» (Very poorly) 11. How was the course workload?35 svarande
Genomsnitt: 4.2 - It was high in a good way. You learn a lot and i dont believe that you will be able to learn as much if the workload wasn"t that high. With lower workload you would simply not get any deep understanding in the topic. » (High) 12. How was the total workload this study period?35 svarande
Genomsnitt: 4.2 - When both courses had deadline in the same week, it was very tough.» (Too high) Summarizing questions13. What is your general impression of the course?35 svarande
Genomsnitt: 3.71 (bidrar till totalt genomsnitt/jämförelseindex) - The course it"s focused on physics and mathematics part of the neural networks, not on the computer science field.» (Poor) 14. What should definitely be preserved to next year?- The examples classes»- The example sheets, the great attitude towards questions, examples class, the litterature and lecture notes. The free choice of programming language. The not too detailed example sheets" instructions (which makes report writing actually mean something), and forces you (in a good way) to seek literature and to think!» - Examination procedure» - The assessment method for this subject is the proper one.» - Bernhard and lecture notes» - The teacher :P» - assignments» - Lecturer, lecture notes, exercises. » - The implementation part of the laborations.» - The examination. » - everything» - Example sheet classes, especially for the first ones when you"ve not yet familiarized yourself with theory or the programming. It can be very hard to understand what is being asked for in the exercises if clear instructions has not been provided.» - Inspiring lectures. Lecture notes. Examples sheets.» - Great lectures» - The Examplesheets this is a very good way of learning. » - The general idea of the course is fine.» - The Examples classes» 15. What should definitely be changed to next year?- 1. Don"t stand in front of the black board while writing and speaking. Speak afterwards instead and it"d be easier to follow. That"s gonna cost you some extra minutes on the clock though :) 2. Maybe some review of the example sheets, i.e. how good can it get on the pima indian set, for instance? What"s the best solution so far on this type of dataset? And analogies to other real world examples. The Krogh book has some neat applications concerning how much weight decay improved a solution, for instance. 3. Reading instructions before lectures would be good, or _at least_ a schedule so that one is prepared of what topics are to be brought up. I have to admit that I lost track of the subject the last few lectures...»- Cooperation between students, available tutoring» - Reference material. The notes were the worst ever.» - Marina. She was beyond acceptance.. She responded some questions in class in a highly disrespect maner.. Who cares if it is a stupid question... She can"t stand there acting as a prit.. She responded emails as "you have to figure it out your selfe, good luck".. She seem"s to completely missed both the "teaching" and "assistance" in her role as teaching assistant... » - Testing different parameters didn"t felt very rewarding, as you were not given enough information on their effect or what you were supposed to get from it. Having weekly exercises that were difficult to implement made you have little to no time to understand what you did or to write about it, also kept me constantly stressed during the course. It was really hard to follow the long equations on the blackboard.» - It was very hard to follow the lectures. Please give an clearer overview. I was last in details and didn"t see the whole image.» - I think the exercise lectures can be developed. They were very ad-hoc in a bad sort of way, often leaving me more confused afterwards than I was before the lecture. What would really help is explaining what the exercise sheet aims to teach and actually going through the questions (but not telling us how to solve them) rather than just hinting at what graphs to draw and numbers to look at. Often I found that Marina just started talking about some questions without really introducing it and with statements such as "this is exactly what you have done before, it should be easy", which if you have read the question and not really understood what part is "exactly like a previous exercise" is just demoralizing. Also, doing easy stuff, generally isn"t much fun...» - The example sheets should have all necessary information on the PM. It is STUPID(!) that you almost have to attend the "walkthrough" at the lecture in order to know what they really want you to do - if these walkthroughs will stay for next year, make sure that the information given also is presented on the PMs! I feel that the workload for the example sheets is not distributed well wrt to coding (problem solving) and writing the report. For me (and all my friends) the coding (ie solving the problems) occupied only a small fraction of the total time spent on producing the reports (of course running time will take a lot of time, but the process of producing figures and data which could be included took too much time i think).» - I don"t understand why ex5part2 is interesting or teaches something.» - * More clarity in the assignment texts! If one is supposed to experiment with diffrent layer-configurations and vary three different variables it"s just a waste to spend five hours varying a parameter between one and ten thousand when its supposed to be in the range 0 to 10^-4.» - Specify a bit more on the example sheets what are expected of you. Espescially ES 3 had way too many paramaters to be varied, with long running simulations. Would be better if specified (in the paper rather than the example sheet classes) which different combinations we should try, or rather what kind of results we want to achieve. Would also like to know how many points needed for grade 4 and 5» - It should be clear the physics and mathematics basis of the course.» - Not 2 lectures in a row, it"s too much for this kind of subject. The lectures contain many deriviations and 4 hours of it is too much.» - The opportunities to ask questions or just discuss anything, really. I believe that having a discussion with someone who knows the topic is a huge boost in how well you understand and learn the course. If you have a lot of thughts and ideas and not get them confirmed/denied by discussing, no new thoughts and ideas can spin off from the original ones. Therefore i would think it would be appropriate with some kind of class where you can work with the examples sheet and discuss the exercises and your ideas how to solv them, with a teacher or exercise leader. I dont know what the purpose of the examples class was this year, but it did not really give as much understanding of the problem as I expected it would give me. Also I think that a little too much weight was put on the theoretical part of all the topics discussed. There was little to no discussion in lectures of how to map a real problem to the different network topologies. The only thing said was that you could use this kind of model to solve, this, this and this problems, but no time was spent in explaining how to actually solve them with the theory we have learnt. A lot of that was left to the student to figure out. Just a simple example on the black board would give so much more understanding in how to use the theory we have just learnt. I remember that during classes long calculations were made to prove for example the convergence of the network. During this long calculations I realized that I have forgot what we were actually trying to prove. You were so focused on following the proof and trying to understand it, that you forgot what you were proving. A good idea might be to pause the calculations more often and remind what we are actually trying to achieve with the calculations.» - Rework the lecture notes to be more clear. Better information for students not attending many lectures.» - Do something about the workload.. but at the same time it is stimulating.. and I really dont know how to change the course since it is a really good one as it is.. » - Example sheet 5. It was conceptually interesting, but in practice mostly repeating things we did already.» - Use standard notations and standard names. Obviously the threshold isn"t called threshold if you use it the way the course uses it, in which case it"s called bias. The symbols for different things also should be the standard ones. It"s not even the same in the course as in the litterature it uses. Make sure to actually answer mails and questions wether the answer is "i can"t tell you because it"s part of the assignment" or a proper answer.» - Three thirsdays in a row we had two classes in a row, i.e. 4 hours. That was way too much» - Perhaps have some introductory math lesson to get everyone on equal footing. Or at least be clear if something that some may have not encountered before is going to be used in a coming lecture, so you can study it yourself. Sometimes we got heads-up what you could look into, but it was usually too late. Trying to understand a lecture that was based on some previously unknown (though not necessarily difficult) mathematical concept is very hard. » 16. Additional comments- The teacher assistant was completely unuseful and in more than one occasion of asking question (by mail also) the answer was (i quote) "i think that the notes are very clear concerning what needs to be done. please see the book as well. Good luck". As a student i am not here to waste my time and the fact that i have red the book and the notes before the question is taken for granted. When i ask a question, for stupid that the question can be to those who know the subject, i would like to be helped by a real answer that does not involve luck as key concept of solving the issues. If we could learn the subject just by reading the book and "engineer" ourself to find a way out, i do not see the need of taking a course.»- Bernhard is great! » - Thanks for an overall great course!» - A really interesting course.» - vielen dank! » - Great course idea. Sadly a bit bad executed idea.» - I"d like Bernhard to play the bassoon on one of the lectures!» Genomsnitt totalt för detta stycke: 3.71 Beräknat jämförelseindex: 0.67
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
5%