connectionist 0.2.5

Tools for classical connectionist models of reading with TensorFlow

Connectionist

Connectionist contains some tools for classical connectionist models of reading in TensorFlow. This project is a course companion python library for Contemporary neural networks for cognition and cognitive neuroscience.

Features

• Ready-to-use models of reading in TensorFlow
• Various "brain" (model) damaging APIs
• Basic building blocks (layers) for connectionist models

Requirements

• Python >=3.8
• TensorFlow >=2.9

Installation

pip install connectionist

Quick start

End-to-end toy example with Plaut, McClelland, Seidenberg and Patterson (1996), simulation 3 model:

import tensorflow as tf
from connectionist.data import ToyOP
from connectionist.models import PMSP

data = ToyOP()
model = PMSP(tau=0.2, h_units=10, p_units=9, c_units=5)
model.compile(
    optimizer=tf.keras.optimizers.Adam(),
    loss=tf.keras.losses.BinaryCrossentropy(),
)
model.fit(data.x_train, data.y_train, epochs=3, batch_size=20)
model(data.x_train)

Documentation

https://jasonlo.github.io/connectionist/