Inference eNgine uSing aPproximate arithmEtic ComponenTs for Neural Networks
Project description
# inspect-nn : simPle sImulator for Neural nEtworks using APProximate muLtipliErs (written in python)
inspect-nn is a simulation tool for measuring the classification accuracy loss due to the use of approximate multipliers (i.e., multipliers designed by exploiting the [approximate computing design paradigm](https://link.springer.com/book/10.1007/978-3-030-94705-7)) in neural networks.
It has been developed to easily do porting of keras/tensorflow models, and to use the GPU to simulate a neural network while using approximate arithmetic units quickly and almost effortlessly.
Anyway, it is still under development!
## Installing the module You can install this module using `pip` `bash $ pip install inspect-nn `
## Using inspect-nn
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.