Welcome to GradForge

Machinelearning library created for full customization of learning process

Models and layers with: • Autograd
• Callbacks • Metrics • Preprocessing functions

Project published on PyPI

Install it with pip

pip install gradforge

Quick Start

from gradforge.core import Tensor
from gradforge.optimizers import SGD
from gradforge.autograd.standard_engine import StandardEngine
from gradforge.backend import BackendManager, NumpyBackend
from gradforge.training.layers.linear import Linear

BackendManager.set_backend(NumpyBackend())

x       = Tensor([[4], [8], [9]])
y       = Tensor([[8], [16], [18]])
test    = Tensor([[40], [60]])

model = Linear(1, 1)
engine = StandardEngine()
optimizer = SGD(learning_rate=1e-3)
for epoch in range(1000):
    y_pred  = model.predict(x)
    error   = y - y_pred
    loss    = error ** 2

    optimizer.zero_grad(model.params)
    engine.backward(loss)
    optimizer.step(model.params)

print(model.predict(test).data)