Data Science Uncut - Data Shootout Kaggle Competition (Aug 1 2022 Stream)
HTML-код
- Опубликовано: 9 авг 2022
- In this Data Science Uncut we look at the Bundesliga Data Shootout kaggle competition. This is a rebroadcast of a twitch stream where we talk data science, machine learning and coding in python.
Follow me on twitch for live coding streams: / medallionstallion_
My other videos:
Speed Up Your Pandas Code: • Make Your Pandas Code ...
Speed up Pandas Code: • Make Your Pandas Code ...
Intro to Pandas video: • A Gentle Introduction ...
Exploratory Data Analysis Video: • Exploratory Data Analy...
Working with Audio data in Python: • Audio Data Processing ...
Efficient Pandas Dataframes: • Speed Up Your Pandas D...
* RUclips: / @robmulla
* Twitch: / medallionstallion_
* Twitter: / medalliondata
* Kaggle: www.kaggle.com/robikscube
#kaggle #python #livestream
Just started with Machine Learning 2 months ago and already amazed with what you can do with ML/AI , you guys are really Online Teacher's are really inspiring a new age of engineers . Thank you so much
Great to hear! These live streams can sometimes be all over the place but I'm glad you find them useful!
it´s so nice to see you decoding... it show us how we must study and the most important be curious and not be afraid to see and work with something which is not familiar to us. Greatings from Brazil!
Absolutely! Thanks for watching.
This stream was soo great. Please keep posting more of such unedited streams, I love it 😍.
Thank you! Will do! You should also check them out live sometime if you can. I've been streaming to twitch & youtube from time to time.
Just found your videos, they are amazing! Thank you for your work :)
Thanks for watching. I appreciate the feedback.
Happy to see these uncut videos also on RUclips!
Glad you like them!
Do more of this as we can learn to start and go through the problem / competition as a beginner. Thanks for these uncut videos
Glad you like it. Thanks for watching!
@@robmulla Hey, also it would be good if you livestream for the recent updates regarding the competition, discuss personal or other's developments. Or else many of us come to dead-end after a point for competition like this. Thanks !!!
deep sort's kalman filter assumes constant velocity and acceleration, it also assumes linear motion, so the ball suddenly changing movements messes with it a lot.
Love this concept ! Thank you ;)
You're so welcome!
Thank you for the content.
Thank you for being you!
For the time of every event you can calculate the difference between "time" and "time" shifted by 2 periods (every event from the start to the end takes 3 rows in the df...)
Interesting like df['time'] - df['time'].shift(2) ?
Time really flies, watching this on 2nd Jan, 2024
Interesting, thank you for your thoughts and insight! FYI, the Twitch streamer Skalskip is doing an ongoing deep dive on this challenge now.
Oh cool. Thanks for watching.
I'm watching from South Korea. it is very helpful. Thanks. Someday I wanna join the streaming.
Awesome! Thanks for watching, hopefully you can join one in the near future!
wow pretty cool…didn’t realize you were on twitch!!
Thanks! I've actually been on twitch longer than I've been making youtube videos!
Hey its really interesting concept to follow along competitions live just found your channel would love be in the journey further. Any guidance for someone who has attended few competitions already and wanna boost rank further.
Thanks for watching. There are a lot of tricks to improve your model but most of the time you just need to test different ideas.
I used this datasets and published a conference abstract.
OMG, you soo expert...
nice video👍👍👍
Thanks for watching! 🙌
ML beginner here, so I have a noob question: why do you prefer to work on jupyter lab instead of doing it on a kaggle notebook? Specially since you have to manually download the data and store in your local storage vs having it accessible in the cloud.
Thanks!
Mainly because the notebook will log off after a while and tends to crash more often. Also my machine is much faster.
Great
What color theme do you use for Sublime? Is it Nord?
Do you mean my jupyter lab theme? It is nord! I installed it from the jupyterlab-themes package. Thanks for watching.
nice video, im a cc student and i very interested in area. Greatings from Brazil!
Awesome, thank you! Glad you are interested in data science. It's a lot of fun!
Can u give some tips for the people who r new in kaggle?
Thanks. I have a video on how to get started with a kaggle competition. I'd suggest checking it out!
very very amazing. please Mr can you share your dataset with me??? it is not available on kaggle
Thanks for watching! The dataset should be available through kaggle here: www.kaggle.com/competitions/dfl-bundesliga-data-shootout/data
@@robmulla once again, thank you very much sir. but it is impossible to download the dataset before december 25th (because the deadlines for joining the competition have expired). in fact, in the video you have made a download of the dataset. That's why i was asking if it's possible to share it
@@robmulla please sir do you have any idea on how to detect shot and scene changes in a video?
Suggesting cutting a bit for RUclips; Or add a timestamp please
I definitely need to do something like that. Or timestamp. I don’t have the time to fully edit these at this point.
all the players are looking at the ball :) 1:36:00
[bu: ndes leega] ;)
Did I pronouce it wrong :D
@@robmulla no worries, it was fun hearing it. :)