🎼Up and down
🎶 And all the way around
🎶 Crossentropy is a bitch
🎶 Epoch and loss are still the same

Edit: I left this rant open for half an hour. Finally figured that I should do one hot encoding and labeling doesn't exactly work out the way I was hoping it works out.

But hey, half an hour ago I was angry and desperate. So... ¯\_(ツ)_/¯

  • 3
    Who da fuq down voted this as first? Bitch, seems like I've got haters here...
  • 2
    @NoMad haters gonna hate.

    Good thing they're a minority 😉
  • 4
    @C0D4 I could make a wild guess of 3-4 people who might do it, although I think 2 of them have already left devrant.

    Either way, good to know that I'm important enough to have haters 💅👄🤳
  • 2
    @NoMad damn girl, I'm jealous of your fan club!
  • 5

    Biggest fucking facepalm.

    I wrote "Epoch" when I meant accuracy....

    Yeah, I'm gonna quit and go dig graves for a living...
  • 1
    So disappointed in myself rn... 😓😞😢😢

    "Dyslexic" doesn't cover the amount of fuck ups I make while talking/writing and meaning one thingy instead of the other thingy.
  • 4
    @NoMad You and me both. Or: I forget names for things so I just describe them instead. I called our dishwasher a “hot water box” yesterday. 🤷🏻‍♀️🤦🏻‍♀️

    Unrelated: I love the vertical emojis!
  • 3
    @NoMad I was going to ask, was extremely confused for a while lol (the epoch thing)
    But eh, no worries
    Screw the haters, this is relatable.
  • 2
    @RememberMe I was simultaneously fiddling with epoch, batches, metrics, optimizer, loss function, and input shape. So yeah kinda got confused.

    I couldn't be fiddling with more things at once even if I were an octopus 😜
  • 1
    @NoMad maybe you can bury me while Im at it. this pandemic and civil unrest shit is gonna put me in my grave from alcoholism.
  • 3
    When you start looking at the accuaracy upon each epoch and you don't see any improvements ;__; this shit is so relatable but it always goes back to the same thing gurl: data preprocessing.

    ahhhh ai......2 minutes writing the model....5 days preparing the data..........ded. Not even understanding the intricacies behind the mathematics is such a pain in the ass.
  • 0
    @AleCx04 oh, I understood the math (or so I think) but I was trying magic with my output tensor. And plus, the noisy data was too big. (there, I said it) so I tried to resize the input too, without losing too much goodies. The accuracy shot above 95%, and the tests came back with interesting results. I don't even have that many samples available. (which is the real drama here)
  • 1
    @Wisecrack nah you'll be fine. Why is it that the majority always panics when the minorities ask for equality? I mean, Unless you're in the dead center of the protest where you may get shot by random militia, you're gonna be fine.
  • 0
    @NoMad because equality is the ideal, but its used as a lie. and what they will get once this is done is something worse in terms of government.
Add Comment