How real is this?

[TarzanEng]

I mean compare to real life? I am no programmer so I do not know how real can we equate this to real life?

Anybody

(Obviously real life is harder but just want to know howclose it could get)

[ScubaSteve]

I am kinda wondering the same thing. I am learning some python and java. Wondering if playing this game has any crossover to languages and such or purely for entertainment and gaming?

[Hallowedsoul08]

i imagine that if it's real, you have to code everything yourself via the powerful language scripts available from the game's engine.

Then this game becomes more of an educational game then it is a puzzle game.

[ScubaSteve]

Hoping someone who has played the game can chime in lol

[ALFonsin]

I will try this right now, brb in 4hs and I tell you

[ScubaSteve]

Originally posted by ALFonsin:

I will try this right now, brb in 4hs and I tell you

Nice

[The Flaming Red]

Well it's not going to teach you any actual written programming if that's what you're asking. It is however, going to give you practice on your logic skill to solve problems on a nice graphical interface. IF statements and FOR loops are the two building blocks of most manipulation programs you'll come across and this game will set up scenarios to use those skills. But in my short time playing so far, it nicely introduces the concept of 'not needing true accuracy' in quite a fun way.

it is in essence, a puzzle game along the line of Human Resource Machine and CargoBot (ios)

[ALFonsin]

I've reached the reinforcement learning so far.

In general

It shows exactly the way that Data Scientists struggles with algorithms and time comsumption. Sometimes, if you really love this, you might get frustrated, and that, my friend, is how you learn.

For example (spoiler alert), in "Errors in Database" you have the logic option to put "If this box is red, then put it, and if it's green, then put it, and if it doesn't go in either, put it in the trash". Instead, you can make a different point of view: "if it's blue, put it in the trash, else, put it". And it's less time consuming.

Real world

It's difficult in the real world for two main reasons:
- Most of the time you pass in the data instead of the algorithm
- It's hard to visualize what you've achieved

For those two reasons, the "solutions":
- Learn Python/R or some language to "clean" the data
- Learn how the models (what we build here) are measured (aka metrics)

To sum up

In general, it's pretty sweet. It really makes you love Data Science and also shows lot of history (which is important, because ML is not new at all)

PS.: All the content that this game suggest, you should TOTALLY try it out.

[ScubaSteve]

Originally posted by ALFonsin:

I've reached the reinforcement learning so far.

In general

It shows exactly the way that Data Scientists struggles with algorithms and time comsumption. Sometimes, if you really love this, you might get frustrated, and that, my friend, is how you learn.

For example (spoiler alert), in "Errors in Database" you have the logic option to put "If this box is red, then put it, and if it's green, then put it, and if it doesn't go in either, put it in the trash". Instead, you can make a different point of view: "if it's blue, put it in the trash, else, put it". And it's less time consuming.

Real world

It's difficult in the real world for two main reasons:
- Most of the time you pass in the data instead of the algorithm
- It's hard to visualize what you've achieved

For those two reasons, the "solutions":
- Learn Python/R or some language to "clean" the data
- Learn how the models (what we build here) are measured (aka metrics)

To sum up

In general, it's pretty sweet. It really makes you love Data Science and also shows lot of history (which is important, because ML is not new at all)

PS.: All the content that this game suggest, you should TOTALLY try it out.

Interesting thanks for the response/review.

[Not a Bard]

I think I'm a little past half way through and ... it's mostly flavor text. The actual machine learning that you do is minimal, and usually something like "plug in this module and click train until it says 20%", then use it like you'd expect. The programs you make don't "learn" in a machine learning way . (Except the car minigame, which does seem to be something you can teach simplistically.)

It's been kinda fun to play, but it's not difficult - the hardest part is deciding how to distribute the amount of a given input to match the expect amount of output. In fact, I feel a little underwhelmed as far as puzzle games go, but that's off topic.

[Phishfood]

So to preface I know some programming but I am by no means elite, or even intermediate.

From what I can tell so far the game gives you medals (bronze/silver/gold) for being able to setup a solution that completes the puzzle within a certain amount of time and with a limited number of nodes.

Now I don't know anything about machine learning but I imagine it's a very CPU intensive task so the game rewards you for being optimal with your solutions.

[El Guapo]

Hmm, I started working with Nintex, Flow and PowerBI recently.

Just stumbled upon this game. Do you think the logic is similar to workflows?

[Belle Delphine Targaryen!]

Originally posted by Priest Of Rathma:

I am kinda wondering the same thing. I am learning some python and java. Wondering if playing this game has any crossover to languages and such or purely for entertainment and gaming?

Greetings, Py bro! Makes me riotously happy when Python is brought up!

[TarzanEng]

Thanks guys. I will try it

[category5]

This is an old thread but it came up in Google page1 rank so I'm adding this entry for anyone else who comes across.

I watched through YouTube videos for some time before purchasing to see the development and from what I can see I would say this game is more of a visual meta-coding game. This is more comparable to Scratch or Code.org or some of the Zachtronics games. Not to dismiss it, but this game looks like alot more fun than actually training models with Jupyter/Pytorch :).

If you are looking for a game where you can play with programming more, I would recommend Greyhack which uses a language similar to Lua and Python. You can create fairly complex scripts and get a feel for working with libraries and control flow, and even create basic websites.