Join devRant
Do all the things like
++ or -- rants, post your own rants, comment on others' rants and build your customized dev avatar
Sign Up
Pipeless API
From the creators of devRant, Pipeless lets you power real-time personalized recommendations and activity feeds using a simple API
Learn More
Search - "expression engine"
-
Can someone explain the philosophy of the "not for me" downvote?
There are many things that are "not for me" in life, but, presumably, this action executes a global downvote on the post/comment---which is pretty much an expression of "not for anyone".
If this action were to train a recommendation engine---so I get recommendations that like-minded people see---then great. But why should that result in a public downvote?
I don't go up to people in the bar and say "Drinking Guinness? Not for me, mate." As an adult, I understand that my preferences are not universal.
Personally, I can't square the idea of "not for me" with its consequence of a public downvote.
I'm sure this must have been covered before...but all rants, as physiological and emotional activities, are unique. Your rant can never be mine.4 -
I built a view engine that relied on V8 for expression evaluation and flow. Not very stable of course, since it used RegEx, but it worked fine for what it was designed for.
The crown feature was the ability to pass in lazy-evaluated huge objects to that view model, so that the view model decided what was going to be used in order to display the view. Made it really flexible, while not sacrificing speed.
I was brainstorming for 2 days about the lazy loading part, and the gymnastics that had to be implemented for this to work.
After I wrote my final line of code and thought that this is it, I launched it, and it FUCKING WORKED! First try!
I was hyperventilating, walking around the apartment like crazy, doing random push-ups just to try to utilize some energy that I felt was fighting to burst out like a xenomorph out of the chest.
... 2 weeks later I found bugs. Had to re-learn how I did it. It's true what they say: if it was hard to write, it's even harder to debug. Fixed it eventually, but that part's not that exciting. -
I recently went through a very detailed and well-explained Python-based project/lesson by Karpathy which is called micrograd. This is a tiny scalar-valued autograd engine and a neural net on top of it.
The project above is, as expected, built on Python. For learning purposes, I wanted to see how such a network may be implemented in TypeScript and came up with a 🤖 micrograd-ts - https://github.com/trekhleb/... repository (and also with a demo - https://trekhleb.dev/micrograd-ts/ of how the network may be trained).
Trying to build anything on your own very often gives you a much better understanding of a topic. So, this was a good exercise, especially taking into account that the whole code is just ~200 lines of TS code with no external dependencies.
The micrograd-ts repository might be useful for those who want to get a basic understanding of how neural networks work, using a TypeScript environment for experimentation.
With that being said, let me give you some more information about the project.
## Project structure
- [micrograd/](https://github.com/trekhleb/...) — this folder is the core/purpose of the repo
- [engine.ts](https://github.com/trekhleb/...) — the scalar `Value` class that supports basic math operations like `add`, `sub`, `div`, `mul`, `pow`, `exp`, `tanh` and has a `backward()` method that calculates a derivative of the expression, which is required for back-propagation flow.
- [nn.ts](https://github.com/trekhleb/...) — the `Neuron`, `Layer`, and `MLP` (multi-layer perceptron) classes that implement a neural network on top of the differentiable scalar `Values`.
- [demo/](https://github.com/trekhleb/...) - demo React application to experiment with the micrograd code
- [src/demos/](https://github.com/trekhleb/...) - several playgrounds where you can experiment with the `Neuron`, `Layer`, and `MLP` classes.
Demo (online)
---------------------
To see the online demo/playground, check the following link:
🔗 https://trekhleb.dev/micrograd-ts3 -
Useless language feature #1: specify kind in explicit expression type annotations that you insert to guide the type inference engine.
How did I work on this for 6 months without realizing that the kind of a value's type is always the kind of types because that's literally what the kind of types means?2