A simple python vector package.
Project description
Vector
A python vector library
Contents
Why vector
Vector is a mathematical library for python. In my search to understand artificial intelligence (AI) I strived to make my own neural network entirely from scratch. To do this I began with numpy – a python library which is the go-to for mathematical things within python. As I started to make this program, I realised that I wanted more control over the objects and wanted to use them in such a way that would make sense for AI. So, I made my own library – vector – a simple vector library in python which holds the basic activation functions for neural networks, random generation, and of course a Vector object which is to be used in a way similar to np.array
.
Usage
To import the library:
from vector import Vector
from vector import activation
from vector import random
To define a vector object, a = Vector([1, 0, 0, 1])
or a = Vector(1)
. This class definition takes in either a list, tuple, int, or float.
To generate a random vector, a = random.random_vector(length, lower_bound, upper_bound)
, lower_bound
and upper_bound
are optional arguments and by default are -0.1 and 0.1 respectively.
The vector library has 2 activation functions and the 2 derivative functions of those functions:
activation.sigmoig(vec)
activation.sigmoid_prime(vec)
activation.linear(vec)
activation.linear_prime(vec)
In the example above, vec is either an integer, float, or Vector. sigmoid
is: 1 / (1 + e ^ -x) and linear
is: 0.5 * x.
The _prime
in the function name means derivative.
Installation
To install this library using pip, you need to type pip install vector-ai-ml
.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for vector_ai_ml-0.0.3-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 226fa6d894eb0ddb8a59f5fde9132716ce3bc19d9dc39d6dbce3f5f7c1603f7c |
|
MD5 | 48446ae8f1f3471f311832708071d5a4 |
|
BLAKE2b-256 | 2553d1eb5e22dad384a4834c9d3d6041282ce52c8495797629f61b0cdfbb5e3d |