Training Loop by DomNomNom
This Game puts you in the shoes of a machine learning optimizer. Go forth you glorious neural network!
Stuck?
Try entering negative numbers for different parameters.
Ratings
| Overall | 1296th | 3.136⭐ | 24🧑⚖️ |
| Fun | 1417th | 2.727⭐ | 24🧑⚖️ |
| Innovation | 37th | 4.25⭐ | 24🧑⚖️ |
| Theme | 759th | 3.659⭐ | 24🧑⚖️ |
| Graphics | 1400th | 2.659⭐ | 24🧑⚖️ |
| Given | 16🗳️ | 21🗨️ |
Maybe a bit more feedback on what altering the parameters actually do may be useful :) As humans we can't iterate thousands of time to figure it out on our own
Usually people can cheat by just clicking next all the time and that ruins the fun. (Maybe level progression when a condition is met?)
(Also a + for somewhat teaching the player)
Pretty much the rest is fine :)
Feel free to add some images of parameters sets that work well.
For this concept to be approachable in any way, it needs a lot more explanation, visualization and helpers.
On the other hand, I think in order to provide a bit more rewarding feel to the game, you could have chosen something different for the training set ( I dunno if that's the right word ), rather than the spinning stick. Something like a walking stickman, or some image dataset. I dunno what I'm saying is even right at all lol. What I mean is something other than the rotating stick could have made the gameplay a lot more rewarding. But I think I am very much underestimating the amount of work to be put into it.
But either way, I liked how different this game was. A very unique entry to the jam! :)