[Meeting Notes CX] SW15

[FloBaumgartner]

The timeline for the discussion:

• roket failed (used to much time)
• lstm as baseline model
• main model is training, but not good
• much achieved with scraping during semester and we are proud
• deadline on sunday

What are the most important insights since last weekly?

No response

Where do we stand in our timeline?

• almost before deadline

What are the most important topics discussed?

• models deliver bad outcomes
• what do we do until the deadline
  • -->minimultirocket to hopefully get a usefull model
• it would be good, if one model would be 'useful'

where are we?

No response

What was decided?

• try sktime minirocket multivariate until deadline
• just one model that works is enough
• deliver paper und GEDP on sunday
• text moritz, when we have an output that is better than random

Feelings:

• overall: stressed but okay 🥲
• project: same same dut diffrent 😄