Feedback - Online Hackathon


#1

If you have any feedback from today’s event, please leave them here. The more specific you can make them, the better it would be. It will help us make these events better.

Once again, thanks for joining the fun.

@Manish, @saimadhup, @kesavkulkarni, @ankurbh07, @qingkai_kong, @imranuddin13, @Jegan_Venkatasamy, @nayan1247, @nramachandran86, @Rudra_Saha, @Prateek_Saxena, @Cross_Bow, @Ankur_Khandelwal, @Sravanth_Kumar, @yakkali_sreenu, @Uday_Bhan_Singh, @Tushar_Kakkar, @Manoj_Hans, @Amartya_Hatua, @kmohan10, @Sunny_Malhotra, @Jayanti_Bhanushali, @Siddharth_Paturu, @gauravkantgoel, @ParmodKumar2, @Sakshi_Dhama, @ankitbhargava06, @al3hIshek, @Pragith_Prakash, @Harshil_Gandhi, @kadusrahul, @Ajoy, @Bruce_N_Sheri_Lyn, @Rizwan_Hudda, @mrajugoud0, @Vivek_Agarwal, @mdebasree9, @joshij, @karthiv, @dkanand86, @Miruna_Popa, @RAM_SINGH, @Arpit_Sidana, @vijayakumar_jawaharl, @gauravkumar37, @Alok_Kumar_Singh, @Keshav_Shrikant, @abhinavunnam, @shubhamgoel27, @mail1, @vikash, @explorer, @Gaurav_Bansal, @guptashubham389, @sagar1785, @jayesh1109, @sumit1, @deepishadwani, @ParindDhillon, @Saranya, @rithwik, @naveen26246, @harshals1, @Hardik, @Inglorious_Ankit, @varunmmm, @Aditya_Joshi, @ashimkapoor, @Ishan, @vijendrapsg, @ranjanpossible, @Debasis_Swain, @Ashish, @devanshrising, @dada_kishore, @Prem_Sangeeth, @Basu, @prakashjhawar, @Atul_Sharma, @Aksgupta123_1, @poornaramakrishnan, @Aman_Srivastava, @Sajan_Kedia, @emohit, @savy2020, @chaitumart, @shuvayan, @kamol1002, @aayushmnit, @BALAJI_SR, @Sunil, @tavish_srivastava, @karthe1, @anon, @kunal, @ajay_ohri, @Aditya_Sharma, @manneabhinay5, @Karthik_Ramasubraman, @koustuv10, @Sambit_Mishra, @vasum, @amarsharma999, @sarthak93, @Malavika, @vishwas_an, @gauravchawla03, @nitish_dydx, @vishesh16, @litankumar, @krishnakumar85, @Nalin, @dagreatemly, @John_Step, @savioseb, @Tushar_Sircar, @Harshita_Dudhe, @Sayantan_Sanyal, @aatishk, @Aditi_Sen, @krypton, @thimmasanidineshredd, @r_marouf, @win_nerds, @Abhishek_Nagarjuna, @abhijit7000, @Swapnil_Sharma, @sbairishal, @palashgoyal1, @Anitha_Ramesh, @Ankit_Dhawan, @RoveR, @dark_knight, @Mark, @alok29, @umeshnrao, @Kalyan, @vishwanath18


Hackathon Problem: Do you know who's a Megastar?
#2

Totally enjoyed. Thanks Kunal and AV team


#3

Excellent hackathon. enjoyed and learnt a lot. eagerly waiting for solutions for the generating the best accuracy. looking forward to more such events.


#4

Hope you guys had fun. I couldn’t participate for two reasons: one, my comp. at home had gone bust and I spent the better part of the weekend re-installing the OS and restoring the files; and two, my short exposure to data analysis has already gone rusty with work taking up a lot of the time. If all goes well I’ll try my hand at this next time 'round. :smile:


#5

Problem was very good & so data. If possible, this competition should be run for lengthy time like Kaggle’s.


#6

Congrats AV team. The event was great. The time limit was a unique factor in this competition.
But some feedback from my side.
This type of dataset and the time limit are inversely correlated.
As you know the dataset demanded 80% of feature engineering. First of all engineering features and creating new important features take time. Secondly when you increase the number of features, the model takes time to run.
If the time limit is fixed for these type of competitions, then the dataset should be different. Though this dataset itself was very interesting to deal with, but I guess there are many beginners in this competition. A person learns best by doing himself. In such a short time, the beginners had very short time to make sense of data and come up with decent accuracy with the models.
From a data scientist point of view, I don’t think he will ever be satisfied with only 50% accuracy of his model. It takes time to come up with an efficient model.
Also some time is taken in fixing up errors, and cross validation.
Secondly, everybody must not be having enough RAM and enough cores to get efficient run time. So for them the time limit obviously was less.


#7

i think points made by @gauravchawla03 are very correct.
i also faced some of the issues mentioned above.
for a mini event like this, i think atleast cleaned dataset would have been better option or time should be increased to 2-3 days instead of 7 hour.

due to less time, forum discussions also had some affect. There was no time to participate in forum discussions. To raise a query & wait for response in forum needs time.


#8

Thanks for the Hackathon… We really enjoyed a lot . hope we will get next one very soon :stuck_out_tongue:


#9

Wonderful event. The problem was very interesting and challenging. Thanks a lot for everyone involved.

I personally felt the time given was less and for me lot of time was spent on preparing the data for the model. And also since it was a sunday, couldn’t devote the entire time. Just a suggestion, if you can make it as a 24 hr event starting in the evening, it could be better.