Larq is an opensource Python library for training neural networks with extremely lowprecision weights and activations, such as Binarized Neural Networks (BNNs)^{1}.
Existing deep neural networks use 32 bits, 16 bits or 8 bits to encode each weight and activation, making them large, slow and powerhungry. This prohibits many applications in resourceconstrained environments. Larq is the first step towards solving this. The API of Larq is built on top of tf.keras
and is designed to provide an easy to use, composable way to design and train BNNs (1 bit) and other types of Quantized Neural Networks (QNNs). It provides tools specifically designed to aid in BNN development, such as specialized optimizers, training metrics, and profiling tools. It is aimed at both researchers in the field of efficient deep learning and practitioners who want to explore BNNs for their applications. Furthermore, Larq makes it easier for beginners and students to get started with the field of efficient deep learning.
Getting Started¶
To build a QNN, Larq introduces the concept of quantized layers and quantizers. A quantizer defines the way of transforming a full precision input to a quantized output and the pseudogradient method used for the backwards pass. Each quantized layer requires an input_quantizer
and a kernel_quantizer
that describe the way of quantizing the incoming activations and weights of the layer respectively. If both input_quantizer
and kernel_quantizer
are None
the layer is equivalent to a full precision layer. This layer can be used inside a Keras model or with a custom training loop.
For a detailed explanation checkout our user guide.
Defining a simple BNN¶
A simple fullyconnected BNN using the StraightThrough Estimator can be defined in just a few lines of code using either the Keras sequential, functional or model subclassing APIs:
model = tf.keras.models.Sequential([ tf.keras.layers.Flatten(), larq.layers.QuantDense(512, kernel_quantizer="ste_sign", kernel_constraint="weight_clip"), larq.layers.QuantDense(10, input_quantizer="ste_sign", kernel_quantizer="ste_sign", kernel_constraint="weight_clip", activation="softmax")])
x = tf.keras.Input(shape=(28, 28, 1)) y = tf.keras.layers.Flatten()(x) y = larq.layers.QuantDense(512, kernel_quantizer="ste_sign", kernel_constraint="weight_clip")(y) y = larq.layers.QuantDense(10, input_quantizer="ste_sign", kernel_quantizer="ste_sign", kernel_constraint="weight_clip", activation="softmax")(y) model = tf.keras.Model(inputs=x, outputs=y)
class MyModel(tf.keras.Model): def __init__(self): super().__init__() self.flatten = tf.keras.layers.Flatten() self.dense1 = larq.layers.QuantDense(512, kernel_quantizer="ste_sign", kernel_constraint="weight_clip") self.dense2 = larq.layers.QuantDense(10, input_quantizer="ste_sign", kernel_quantizer="ste_sign", kernel_constraint="weight_clip", activation="softmax") def call(self, inputs): x = self.flatten(inputs) x = self.dense1(x) return self.dense2(x) model = MyModel()
Installation¶
Before installing Larq, please install:
 Python version
3.6
or3.7
 Tensorflow version
1.14
,1.15
or2.0.0
:pip install tensorflow # or tensorflowgpu
You can install Larq with Python's pip package manager:
pip install larq

Hubara, I., Courbariaux, M., Soudry, D., ElYaniv, R., & Bengio, Y. (2016). Binarized Neural Networks. In Advances in neural information processing systems (pp. 41074115). ↩