All Projects → s974534426 → easytorch

s974534426 / easytorch

Licence: MIT license
基于Python的numpy实现的简易深度学习框架,包括自动求导、优化器、layer等的实现。

Programming Languages

Jupyter Notebook
11667 projects
python
139335 projects - #7 most used programming language

Projects that are alternatives of or similar to easytorch

autograd-gamma
NotImplementedError: VJP of gammainc wrt argnum 0 not defined
Stars: ✭ 15 (-80.26%)
Mutual labels:  autograd, autodiff
djl
An Engine-Agnostic Deep Learning Framework in Java
Stars: ✭ 3,080 (+3952.63%)
Mutual labels:  autograd
Pytorch
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Stars: ✭ 52,811 (+69388.16%)
Mutual labels:  autograd
Djl
An Engine-Agnostic Deep Learning Framework in Java
Stars: ✭ 2,262 (+2876.32%)
Mutual labels:  autograd
Flashlight
A C++ standalone library for machine learning
Stars: ✭ 4,038 (+5213.16%)
Mutual labels:  autograd
Pytorch Book
PyTorch tutorials and fun projects including neural talk, neural style, poem writing, anime generation (《深度学习框架PyTorch:入门与实战》)
Stars: ✭ 9,546 (+12460.53%)
Mutual labels:  autograd
Megengine
MegEngine 是一个快速、可拓展、易于使用且支持自动求导的深度学习框架
Stars: ✭ 4,081 (+5269.74%)
Mutual labels:  autograd
Hetu
A high-performance distributed deep learning system targeting large-scale and automated distributed training.
Stars: ✭ 78 (+2.63%)
Mutual labels:  autograd
eddl
European Distributed Deep Learning (EDDL) library. A general-purpose library initially developed to cover deep learning needs in healthcare use cases within the DeepHealth project.
Stars: ✭ 31 (-59.21%)
Mutual labels:  autograd
pytorch-examples-cn
用例子学习PyTorch1.0(Learning PyTorch with Examples 中文翻译与学习)
Stars: ✭ 54 (-28.95%)
Mutual labels:  autograd
Tangent
Source-to-Source Debuggable Derivatives in Pure Python
Stars: ✭ 2,209 (+2806.58%)
Mutual labels:  autodiff
Torsten
library of C++ functions that support applications of Stan in Pharmacometrics
Stars: ✭ 38 (-50%)
Mutual labels:  autodiff
Nabla.jl
A operator overloading, tape-based, reverse-mode AD
Stars: ✭ 54 (-28.95%)
Mutual labels:  autodiff
autodiff
A .NET library that provides fast, accurate and automatic differentiation (computes derivative / gradient) of mathematical functions.
Stars: ✭ 69 (-9.21%)
Mutual labels:  autodiff
autodiff
200行写一个自动微分工具
Stars: ✭ 37 (-51.32%)
Mutual labels:  autodiff
buildTensorflow
A lightweight deep learning framework made with ❤️
Stars: ✭ 28 (-63.16%)
Mutual labels:  autodiff
sunode
Solve ODEs fast, with support for PyMC
Stars: ✭ 67 (-11.84%)
Mutual labels:  autodiff
autoptim
Automatic differentiation + optimization
Stars: ✭ 102 (+34.21%)
Mutual labels:  autodiff
MissionImpossible
A concise C++17 implementation of automatic differentiation (operator overloading)
Stars: ✭ 18 (-76.32%)
Mutual labels:  autodiff

easytorch

使用Python的numpy实现的简易深度学习框架,API与pytorch基本相同,实现了自动求导、基础优化器、layer等。

1 文档目录

1. 自动求导基础运算实现

2. 优化器实现

3. 损失函数

4. layer的实现

2 Quick Start

from easytorch.layer import Linear, Tanh, Sequential
from easytorch.optim import SGD
import easytorch.functional as F

# Create a model, optimizer, loss function
model = Sequential(
    Linear(1, 5),
    Tanh(),
    Linear(5, 1)
)
opt = SGD(model.parameters(), lr=3e-4)
loss_fn = F.mse_loss

# train the model
for epoch in range(epochs):
    pred = model(x)
    loss = loss_fn(pred, y)
    opt.zero_grad()
    loss.backward()
    opt.step()

3 Example

  1. 使用神经网络近似三角函数
  2. 使用神经网络预测波士顿房价
Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].