@mineproness/tensorloop

A Library that Supply AI Tools and Large Language model Development tools like Embedding, Linear, ReLU, softmax etc...

Feature

• Embedding Vectors.
• Linear Weights and Bias.
• Softmax forward and Backward.
• CrossEntroy for Loss Calculation.
• Relu for Stable Layer.
• Tokenizer for convert into tokens.

Configs

You can Pass a Object in a embedding and linear function. like :-

import {Embedding , linear} from '@mineproness/tensorloop'

const embedding = new Embedding(1024, 2000, {
    save: {
        filename: "./model/vectors.bin"
    }
})
const linear = new Embedding(1024, 2000, {
    save: {
        filename: ["./model/Weights.bin" , "./model/Bias.bin"]
    }
})

Linear Layer need a filename array because it need to save the Weights and Bias and always use bin extension.

Optimizations

It is Highly Fast Because of Float 32 Array and it is can handle 1024 embedding Size. We build that with Raw Javascript for Best Optimization.

Tokenizer

It is main use to Split Words and Convert into token ids that can be again Deconvert.

like this Exmple

import { Tokenizer } from '@mineproness/tensorloop'

const tokenizer = new Tokenizer()
const text = "Hello, Wellcome to My Channel."

const { tokenIDs } = tokenizer.encoder(text) // ["hello", "$12" , "well" , "come", "$130"....]
const decoded = tokenizer.decoder(tokenIDs) // Hello, wellcome to my channel.

Suggestion

I recoommeded that use TypeScript for Type Safety and know what function is Exist in classes.

And Always Pair with Our tools in Our Package like

import {Tools} from '@mineproness/tensorloop'

const tools = new Tools()

console.log(tools.betterVector(embeedingVectors))

Thanks for Visiting this Site

If You watching this, Please give me a star in Github.