CLI toolkit for Flow language, and integration with python.
Project description
Flow toolkit
A simple development toolkit for streamlining ML and DL workflows.
Getting started
Installation
The base compiler can be installed via pip
:
pip install flow-toolkit
Next, install a plugin for your intended output. For example:
flow -i https://github.com/StealthyPanda/
Any github repository can be used as a valid flow plugin, as long as it contains a plugin.py
in its root directory, and contains a main
function.
Quick Start
A simple flow for a dense neural network would be:
// example.fl
flow linear(x) [weights, biases] {
return
(weights @ x) + biases;
}
flow NeuralNetwork (x) {
let linear l1;
let linear l2;
y = l1(x);
y = l2(y);
return y;
}
build NeuralNetwork simple {
x => 784;
output => 10;
}
Build the flow to a pytorch model with:
flow -f example.fl -o example
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
flow_toolkit-0.0.1.tar.gz
(22.0 kB
view hashes)
Built Distribution
Close
Hashes for flow_toolkit-0.0.1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | e2003a5d23662987c739dcd3a5dfb75054adf39ab1e9680b01396c5c4ef6fe33 |
|
MD5 | 0e510ae3f0f37bb7e9085fbd7616b584 |
|
BLAKE2b-256 | 209adc2660466094d404e869db06741f7c0ae72dde873291066b9fb75ceb9ec3 |