Skip to main content

High level GPU simulations of low level device characteristics in ML algorithms

Project description

Welcome to LowPy (Pre-release)!

LowPy is a high level GPU simulator of low level device characteristics in machine algorithms. It seeks to streamline the investigation process when considering memristive and other novel devices for implementing a machine learning algorithm in hardware. By using the familiar Keras syntax, it will be second nature to write GPU-optimized code to push your algorithm to its limits.

Features

The aim is to focus first on the algorithms most published on in the field of neuromorphic computing, for both static and time series datasets.

Datasets

  • MNIST

Algorithms

  • Single Layer Perceptron (SLP)
  • Multi-Layer Perceprton (MLP)

Activation Functions

  • Sigmoid

Optimization Functions

  • Stochastic Gradient Descent (SGD)
  • SGD with Momentum

Initialization Distributions

  • Uniform
  • Normal

Device Characteristics

  • Write Variability

Requirements

The following are required to use LowPy:

  • GPU: NVIDIA
  • OS: Linux (should work on Windows, not tested)
  • Python 3.0 or newer
  • PyCUDA

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

lowpy-0.1.1.tar.gz (6.2 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

lowpy-0.1.1-py3-none-any.whl (7.2 kB view details)

Uploaded Python 3

File details

Details for the file lowpy-0.1.1.tar.gz.

File metadata

  • Download URL: lowpy-0.1.1.tar.gz
  • Upload date:
  • Size: 6.2 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.3.0 pkginfo/1.7.0 requests/2.22.0 setuptools/51.3.3 requests-toolbelt/0.9.1 tqdm/4.56.0 CPython/3.8.5

File hashes

Hashes for lowpy-0.1.1.tar.gz
Algorithm Hash digest
SHA256 4701ede5c43924a77cecd668cd9d967232574ef5b61d2a3d63827a3dc0694807
MD5 00baebc809f44a2dba1916b13f806d5e
BLAKE2b-256 e9064d94338255772a6ce67d0d464df5257b36799a886f3950c01b19c26a7d8d

See more details on using hashes here.

File details

Details for the file lowpy-0.1.1-py3-none-any.whl.

File metadata

  • Download URL: lowpy-0.1.1-py3-none-any.whl
  • Upload date:
  • Size: 7.2 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.3.0 pkginfo/1.7.0 requests/2.22.0 setuptools/51.3.3 requests-toolbelt/0.9.1 tqdm/4.56.0 CPython/3.8.5

File hashes

Hashes for lowpy-0.1.1-py3-none-any.whl
Algorithm Hash digest
SHA256 c94b6968388b0828b7ee52f9f6d3f56a211d34c38b523ae92b7be5ce83707d7a
MD5 8eda5d0070ba9c1dc73d778ee033be67
BLAKE2b-256 8e0a0f34a9d25fb3b3ca0488cb5cb6cad20dcedf730b70cc58244663c47c67aa

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page