Search Packages

nvidia-cutlass-dsl

CUDA Templates and Python DSLs for High-Performance Linear Algebra

nvidia-cutlass-dsl-libs-base

CUDA Templates and Python DSLs for High-Performance Linear Algebra

nvidia-cutlass-dsl-libs-cu13

CUDA Templates and Python DSLs for High-Performance Linear Algebra

nvidia-cutlass

CUDA Templates and Python DSLs for High-Performance Linear Algebra

dandelion

A light weight deep learning framework, on top of Theano, offering better balance between flexibility and abstraction

sparseml

Libraries for applying sparsification recipes to neural networks with a few lines of code, enabling faster and smaller models

torch-geopooling

The geospatial pooling modules for neural networks in PyTorch

lasagne

Lightweight library to build and train neural networks in Theano

nutsml

Flow-based data pre-processing for deep learning

shinnosuke

A keras-like API deep learning framework,realized by Numpy only.Support CNN, RNN, LSTM, Dense, etc.

whiteboxlayer

TensorFlow based custom layers

nets

Recreating PyTorch from scratch, using Numpy. Supports FCN, CNN, RNN layers.

torchhandle

TorchHandle makes your PyTorch development more efficient and make you use PyTorch more comfortable

kd-lib

A Pytorch Knowledge Distillation library for benchmarking and extending works in the domains of Knowledge Distillation, Pruning, and Quantization.

deepthink

A deep learning library created from scratch

tinynn

A lightweight deep learning library

n3d3

N2D2 is an open source CAD framework for Deep Neural Network simulation and full DNN-based applications building.

dougnet

A lightweight, deep learning library written in pure Python

semiflow

SemiFlow is a deep learning framework with automatic differentiation and automatic shape inference, developing from Numpy. 一个基于Numpy支持自动求导的深度学习框架

starttf

A tensorflow batteries included kit to write tensorflow networks from scratch or use existing ones.

sunstreaker

源码清晰明了，使用简单好搞

n2d2

N2D2 is an open source CAD framework for Deep Neural Network simulation and full DNN-based applications building.