When the course guarantor is selected, comments will be displayed regarding his/her teaching performance as well as that of all the other teachers teaching the course. If a teacher who is not the course guarantor is selected, only comments related to his/her teaching will be displayed.
Mgr. Martin Popel, Ph.D. [32-UFAL], Natural language processing on computational cluster [NPFL118, cvičení]
Student has disallowed display of personal data, 04.01.2024, 1. Year, Computational linguistics, doctoral
Martin explained everything really well and took his time to answer questions. Without him, half of my accounts probably wouldn't work. His tour of the MS building was great.
Mgr. Rudolf Rosa, Ph.D. [32-UFAL], Natural language processing on computational cluster [NPFL118, cvičení]
Student has disallowed display of personal data, 04.01.2024, 1. Year, Computational linguistics, doctoral
Ruda explained everything really well and took a lot of time to answer additional questions and talk about his own PhD experience.
RNDr. Milan Straka, Ph.D. [32-UFAL], Deep Learning [NPFL114, cvičení]
Student has disallowed display of personal data, 29.05.2023, 1. ročník, Computational linguistics, doctoral
There was a great involvement in the practicals provided by our teacher and he was open to a lot of questions on Piazza, technical with debugging or theoretical. He also took some time to help us debug certain issues which is highly appreciated.
RNDr. Milan Straka, Ph.D. [32-UFAL], Deep Learning [NPFL114, přednáška]
Během přednášek je ze Straky přímo cítit jeho nadšení pro věc. Energeticky poskakuje před tabulí jako jeho zvířecí druh. Přednáška je dobře připravená, srozumitelná, Straka vítá dotazy. Ze všech přednášek jsou záznamy.
Student has disallowed display of personal data, 23.07.2023, 1. Year, Informatika - Softwarové systémy, doctoral
Straka je výborný přednášející a velmi dobře se poslouchá.
Student has disallowed display of personal data, 20.07.2023, 1. Year, Informatika - Umělá inteligence, Master's (post-Bachelor)
Milan Straka is my favorite teacher by far. He is always energetic and does amazing job explaining the material. He always replies quickly to the questions of the students and is just pure gold. You can truly feel from his presence that he cares about teaching us as much as he can and preparing us for the world of machine learning.
Student has disallowed display of personal data, 06.07.2023, 3. Year, Informatika, Bachelor's
Vyucujici je megafrajer
Student has disallowed display of personal data, 21.06.2023, 1. ročník, Matematika pro informační technologie, Master's (post-Bachelor)
Naprosto výborný, vždy zapálený přednášející, odhodlaný udělat vše k vylepšení průběhu předmětu, což lze na kvalitě kurzu naprosto zásadně znát. Hodnotím jako nejspíše vůbec nejlepšího vyučujícího, s kterým jsem se při studiu setkal.
RNDr. Milan Straka, Ph.D. [32-UFAL], Natural language processing on computational cluster [NPFL118, cvičení]
Student has disallowed display of personal data, 04.01.2024, 1. Year, Computational linguistics, doctoral
Milan explained everything really well and took his time to answer questions and help with the assignments. His Slurm reference helped a lot with my first experiments on the CPU and GPU clusters.
RNDr. Milan Straka, Ph.D. [32-UFAL], Deep Learning [NPFL138, cvičení]
Student has disallowed display of personal data, 25.06.2024, 1. Year, , Lifelong learning programme - Interest-oriented
Jeden z nejlepších učitelů, co jsem zatím na VŠ potkal, asi nemám moc co dodat.
RNDr. Milan Straka, Ph.D. [32-UFAL], Deep Learning [NPFL138, přednáška]
Student has disallowed display of personal data, 15.09.2024, 2. Year, Informatika - Umělá inteligence, Master's (post-Bachelor)
Jedna z najlepsich prednasok na MFF.
Student has disallowed display of personal data, 28.06.2024, 1. ročník, Matematika pro informační technologie, Master's (post-Bachelor)
The tempo of the course is quite fast and some of the problems are challenging and time consuming(mostly in the end of the semester). It is possible to do this course with other 4-5 courses, but they must be super easy without additional work during the semester
Professor Straka is exceptionally intelligent. He is a menacing combination of strong theoretical foundations and impressive practical skills. His work ethic is admirable; he offers consultation time after his 2-3 hour lectures and often spends the entire day helping students. He is also very active on our student communication channel, Piazza. Although I don't encourage staying up late, he has often reviewed my pull requests at midnight or later, reflecting his relentless dedication.
As a lecturer, he is engaging and animated, encouraging questions and weaving a narrative through his lectures. Although he can sometimes move too quickly, this is likely due to the extensive content he must cover in a limited time. Personally, I learn best through the repetition of new concepts and an intuition-focused approach, which then fuels a deeper dive into the formulas. The course structure could potentially benefit from incorporating more of this approach within the constraints of time and content.
It's been a pleasure and an honor to be Professor Straka's student.
Student has disallowed display of personal data, 14.06.2024, 2. ročník, Algebra, teorie čísel a matematická logika, doctoral
Amazing lecturer, I cannot image the time spent on the preparation of both the lectures and the practicals.
Student has disallowed display of personal data, 04.06.2024, 4. ročník, Informatika, Bachelor's
Milan Straka je bůh!
Student has disallowed display of personal data, 29.05.2024, 3. Year, Informatika, Bachelor's
Milan Straka je bezpochyby nejvíce inspirativním pedagogem, se kterým jsem se ve svém celém životě setkal.
Ať už jsem měl jakýkoliv problém vždy mi dokázal pomoct a i po několika hodinách konzultací s jinými studenty pořád přistupoval k problému s neutuchajícím nadšením a energií.
Zároveň je opravdovou kapacitou ve svém oboru, stíhá sledovat vývoj ve strojovém učení, takže to co se učíme je opravdu aktuální a přímo použitelné v praxi.
Chtěl bych Vám tímto ještě jednou poděkovat za Vaši práci a přínos, který to pro nás má. Děkuji Vám!
RNDr. Milan Straka, Ph.D. [32-UFAL], Deep Reinforcement Learning [NPFL139, cvičení]
Student has disallowed display of personal data, 14.09.2024, 3. Year, Informatika - Umělá inteligence, Master's (post-Bachelor)
Milan Straka is an amazing teacher—very enthusiastic and always willing to answer questions and offer support with assignments. In terms of teaching quality, he far surpassed what I thought was possible. Personally, taking his classes had a profound effect on me. I would not be the same person I am today had I not met him.
RNDr. Milan Straka, Ph.D. [32-UFAL], Deep Reinforcement Learning [NPFL139, přednáška]
Author of entry is known to survey organiser, 28.06.2024, 4. ročník, Informatika, Bachelor's
Velmi příjemný, snaživý a vstřícný vyučuijící.
Pokud se můžete zapsat na jakýkoliv jeho předmět, vřele doporučuji. :)
Comment on course, Deep Learning [NPFL114, cvičení]
Student has disallowed display of personal data, 23.07.2023, 1. Year, Informatika - Softwarové systémy, doctoral
Úkoly na hluboké učení jsou poměrně časově náročné pro jednotlivce, nicméně, je dovoleno (a naopak je to doporučeno) je plnit v malých týmech. Úkoly pokrývají víceméně celou látku probíranou na přednáškách a pomáhají ji lépe pochopit.
Student has disallowed display of personal data, 30.06.2023, 1. ročník, Computer Science - Artificial Intelligence, Master's (post-Bachelor)
The practicals were really helpful to solve the assignments. I have a suggestion : it may be great to make a small tutorial for AIC a the beginning of the year (because, having never used remote clusters, and very rarely the terminal, it was quite indimidating at the beginning and I didn't use it for a long time : I did it only once a fellow student taught me the basic commands, which is a pity).
Comment on course, Deep Learning [NPFL114, přednáška]
Straka je boží vyučující. Na všech jeho předmětěch je vidět jeho zápal pro AI. Materiály k výuce jsou srozumitelné. Pokud něco nepochopíte, stačí se zeptat. Straka odpovídá na zprávy rychle a srozumitelně.
Předmět jako takový je časově náročný, avšak mi dal opravdu hodně. Můžu jedině doporučit!
Student has disallowed display of personal data, 23.07.2023, 1. Year, Informatika - Softwarové systémy, doctoral
Předmět je každý rok aktualizovaný, aby pokrýval nejnovější technologie v oblasti hlubokého učení. Každá přednáška je doplněna jednoduchými úkoly, na kterých je probíraná látka ukázaná v praxi.
Student has disallowed display of personal data, 20.07.2023, 1. Year, Informatika - Umělá inteligence, Master's (post-Bachelor)
This is an incredibly valuable course. The slides are very high quality and the explanations of the teacher are great. The material is modern and reflects current state-of-the-art techniques. After attending this course I feel like I could do anything; whether it's image recognition, text processing like ChatGPT, or image generation like stable diffusion. It is undoubtedly the best course I ever attended.
Student has disallowed display of personal data, 06.07.2023, 3. Year, Informatika, Bachelor's
najlepsi predmet na deep learning na univerzite
Student has disallowed display of personal data, 30.06.2023, 1. ročník, Computer Science - Artificial Intelligence, Master's (post-Bachelor)
It was great to have a teacher that was both passionate, and went in depth but always being clear. The material available (like recordings) was also very nice. My only concern with this course is that doing the assignments, even in team, really takes a lot of time.
Student has disallowed display of personal data, 21.06.2023, 1. ročník, Matematika pro informační technologie, Master's (post-Bachelor)
Velmi přínosný, zábavný a neuvěřitelně propracovaný kurz. Obecně jistě jeden z nejlepších předmětů, které jsem zažil. Vše je domyšleno do posledního detailu se záměrem aby si student odnesl co nejvíce. Nemám vůbec co vytknout, pouze je však dobré počítat s tím, že je předmět poměrně výrazně časově náročný.
Student has disallowed display of personal data, 29.05.2023, 1. ročník, Computational linguistics, doctoral
Great lecture, extremely well documented from papers/articles to implementation strategies.
Comment on course, Natural language processing on computational cluster [NPFL118, cvičení]
Student has disallowed display of personal data, 04.01.2024, 1. Year, Computational linguistics, doctoral
Generally, a good and detailed introduction to the UFAL infrastructure and pointers to important resources. Almost no questions were left open after the 3 days, but maybe it is too detailed.
2 entire days were spent reading Wiki pages, which was tiring. However, PhD students should be able to understand most of the sections by themselves. For example, an additional meeting was supposed to be scheduled for Biblio, but the system is self-explanatory and everyone finished the checklist anyway. I suggest putting the greatest focus on the Linux network and the topics related to the checklist while going into less detail otherwise. Perhaps, some of the more self-explanatory sections can be left as homework, and a Q&A meeting can be scheduled instead of a reading meeting.
The Slurm and Spark assignments were quite fun, but the grading was a bit too nitpicky at times. Since I immediately started my own PhD-topic-related experiments on the cluster, I would have liked to scrap 1 or 2 of the assignments because they took more time than necessary without giving me a real benefit.
Comment on course, Deep Learning [NPFL138, cvičení]
Student has disallowed display of personal data, 25.06.2024, 1. Year, , Lifelong learning programme - Interest-oriented
Osobně jediné, co mě štvalo, byly chvíle, kdy se v domácích úkolech muselo použít něco hodně speciálního a horko těžko zdokumentováno, a nebyli jsme před tím dostatečně varováni - add_tokens() v reading comprehension, detach v ganech a mnohé další - na těchto blbostech se propálilo zbytečně moc času a ano, k naučení je třeba si něco protrpět, tyhle věci mi ale konkrétně přišly zbytečné. Pokud se bude další semestry kompilovat, tak se to nezlepší.
Student has disallowed display of personal data, 14.06.2024, 2. ročník, Algebra, teorie čísel a matematická logika, doctoral
Even though the problems (and especially the competitions) were often times highly non-trivial and took a substantial amount of time, I think that nothing else would have helped me to understand the material that well. Most importantly, all the tasks were accompanied by rather detailed templates with a lot of commentaries, and so at no point I felt totally lost. Finally, the lecturer and additional instructors were always there to help with some particular problems.
Comment on course, Deep Learning [NPFL138, přednáška]
Student has disallowed display of personal data, 28.06.2024, 3. Year, Matematické a počítačové modelování ve fyzice, Master's (post-Bachelor)
Myslím, že moc povedený ale docela náročný předmět. Během jednoho semestru se projdou všechny základní nejdůležitější architektury neuronových sítí. Pan doktor Straka je skvělý, energetický učitel. Před tabulí někdy doslova skáče a může působit trochu jako blázen, ale určitě se mu daří studentům předat nadšení. Předmět je časově hodně náročný. Připravte se na to, že to bude ta hlavní věc, co budete během semestru dělat, ale myslím, že to za to stojí. Na některém úkolů můžete být zasekly i několik dnů. Moc oceňuji Strakovu ochotu pomáhat a odpovídat na otázky na piazze. Dvakrát jsem za ním přišel po hodině s notebookem a asi půl hodiny se se mnou koukal do kódu a pomohl mi najit chybu.
Student has disallowed display of personal data, 28.06.2024, 1. ročník, Matematika pro informační technologie, Master's (post-Bachelor)
This course is a masterpiece. Professor Milan Straka and his team are doing great job especially in helping other students with their bugs and errors. I would say that this course was bit easier for me than machine learning, because you don't do as much manual coding.
"The structure and preparation behind delivering the course are admirable, comparable to the Harvard CS50 course, though without the high production value. The course covers state-of-the-art, cutting-edge, current research work and traces the evolution of this research.
Personally, I found the course content overwhelming to manage on my own. I couldn't fully take advantage of everything the course offers in just one semester. However, the course design supports success through teamwork and bonus points. Despite this, I would have preferred more time to fully digest the material. To provide some context, my total workload this semester was equivalent to about nine 5-credit courses. I believe students with lighter course loads could benefit more from this course. If the course were split into two parts over two semesters, I might have gained more from it. Additionally, this could allow for more credits, which is something students feel the course deserves.
That said, the course is absolutely stellar, up-to-date, and highly practice-oriented.
Student has disallowed display of personal data, 14.06.2024, 2. ročník, Algebra, teorie čísel a matematická logika, doctoral
Possibly the best lecture/seminar I have ever encountered while studying at MFF. The presentation covers both the theory and the practical part more than sufficiently. And most importantly, it all is up-to-date, which is pretty unbelievable taking into account the rate at which the area is currently developing. I highly recommend this subject to anyone interested in deep learning, the amount of knowledge gained after is simply astounding.
Student has disallowed display of personal data, 04.06.2024, 4. ročník, Informatika, Bachelor's
Úžasný předmět, člověk se toho tunu naučí!
Student has disallowed display of personal data, 29.05.2024, 3. Year, Informatika, Bachelor's
Původně jsem chtěl napsat, že se tento předmět řadí mezi nejlepší předměty na Matfyzu, ale to bych mu křivdil. V mých očích je to s odstupem absolutní jednička na Matfyzu.
Jeho obrovský přesah dokazuje fakt, že jsem se s informacemi z něj setkal v mnoha dalších předmětech a naučil jsem se v něm o umělé inteligenci víc než ve všech ostatních předmětech dohromady.
Opravdu nevím, co bych na tomto předmětu zlepšil – díky Milanovi Strakovi je to dokonalost sama o sobě.
Comment on course, Deep Reinforcement Learning [NPFL139, cvičení]
Student has disallowed display of personal data, 14.09.2024, 3. Year, Informatika - Umělá inteligence, Master's (post-Bachelor)
As expected, this was a highly rewarding class. The material was up-to-date and taught with great explanations and demonstrations. The assignments and competitions were really rewarding and resulted in great and practical understanding, while also being very enjoyable.
I also really appreciate how everything is taught in the context of research, with references to research papers and showing their results. This way of teaching really let me understand what we are learning in relation to current research progress.
Overall, this class significantly elevated my knowledge and abilities and is my favorite class so far.
Comment on course, Deep Reinforcement Learning [NPFL139, přednáška]
Author of entry is known to survey organiser, 28.06.2024, 1. ročník, , Lifelong learning programme - Interest-oriented
Předmět je udělán skvěle. Jeden z nejlepších, které jsem kdy měl.
Jelikož je to první běh předmětu s 8 kredity, přibyly nějaké věci, nějaké věci se nestihly, ale pan Straka vždy reagoval vstřícně, flexibilně.
Jediné, co by mě napadlo, že by šlo zlepšit je dodělat úlohy k tématům jako Rainbow a jiným věcem, které jsou nové a jejich výklad byl poměrně dost abstraktní. Ale to bylo v plánu, akorát se to nestihlo. Celkově by prostě bylo fajn mít alespoň jednu úlohu ke každé přednášce.
Author of entry is known to survey organiser, 28.06.2024, 4. ročník, Informatika, Bachelor's
Podle mého předmět velice krásně popisuje problematiku zpětnovazebného učení a dovoluje si fyzicky osahat i implementační detaily konkrétních algoritmů.
Skvělý a výborně vyučovaný předmět. :)
