Last week we experimented with another new format, we held our first interactive group project session at One Identity. Thanks One Identity, and our host Balázs Antal for the venue and generous catering.
This time, similarly to the previous occasion when Dóri lead the session, we held the workshop in English. We now posted the challenge in advance, so the participants could prepare and work on their on solutions and ideas. During the workshop, we started with a general intro about the structure and aim of these challenges in general, then we continued with talking about the issues, obstacles and revelations we encountered.
After a short break, we had some excellent discussions about how can one calculate AUC in special cases, like for a multi-label classification and how to utilize accuracy score within a grid search to evaluate model performance and much more.
While the format was novel for everyone and we even run into some technical difficulties, thanks to the input and active participation of our enthusiastic group members, we had a lively session where we learnt some new things while helping each other through the impasse. We gathered new ideas how to improve our machine learning models further and several exciting discussions were still ongoing even during the pizza session.
This was the Budapest.py Workshop for 2019, thanks for joining us and we hope we will meet again in 2020.
Last but not least, this week we will have our last meetup of the year, don’t forget to sign up here: https://www.meetup.com/budapest-py/events/266361074/
As always, the notebooks are on our GitHub:
You can join us on our meetup page: