Something between Tinygrad and Micrograd
Project description
Robingrad
Something between Tinygrad and Micrograd.
Installation
To install the current release,
pip install robingrad==0.0.4
From source
git clone https://github.com/marcosalvalaggio/robingrad.git
cd robingrad
./build.sh
Examples
- In the examples folder, you can find examples of models trained using the Robingrad library.
- A declaration example of an MLP net using Robingrad:
from robingrad import Tensor, draw_dot
import robingrad.nn as nn
class RobinNet:
def __init__(self):
self.l1 = nn.Linear(5,16)
self.l2 = nn.Linear(16,1)
def __call__(self, x):
x = self.l1(x)
x = x.relu()
x = self.l2(x)
return x
model = RobinNet()
res = model(X_train[0].reshape((1,5)))
draw_dot(res)
Todos
- Implement the
optim
module. - Test for Linear layer.
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
robingrad-0.0.4.tar.gz
(9.3 kB
view hashes)
Built Distribution
Close
Hashes for robingrad-0.0.4-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 306c9936513daa96f46bda5447139438c6b0332d2c58a070200a5c3071576f1c |
|
MD5 | d0fd33f1bf0122756c434d42030cd3f1 |
|
BLAKE2b-256 | a0f00e9ef7a7d5c8f5872d49d03dfc3c4d166ade2e6cce7b37984eb046845493 |