yizt / Numpy_neural_network
Licence: apache-2.0
仅使用numpy从头开始实现神经网络,包括反向传播公式推导过程; numpy构建全连接层、卷积层、池化层、Flatten层；以及图像分类案例及精调网络案例等,持续更新中... ...
Stars: ✭ 339
Projects that are alternatives of or similar to Numpy neural network
Pytorch Image Classification
Tutorials on how to implement a few key architectures for image classification using PyTorch and TorchVision.
Stars: ✭ 272 (-19.76%)
Mutual labels: jupyter-notebook, cnn, vgg
Food Recipe Cnn
food image to recipe with deep convolutional neural networks.
Stars: ✭ 448 (+32.15%)
Mutual labels: jupyter-notebook, cnn, vgg
Pytorch Sentiment Analysis
Tutorials on getting started with PyTorch and TorchText for sentiment analysis.
Stars: ✭ 3,209 (+846.61%)
Mutual labels: jupyter-notebook, cnn
Rethinking Numpyro
Statistical Rethinking (2nd ed.) with NumPyro
Stars: ✭ 225 (-33.63%)
Mutual labels: jupyter-notebook, numpy
Deeplearning cv notes
📓 deepleaning and cv notes.
Stars: ✭ 223 (-34.22%)
Mutual labels: jupyter-notebook, numpy
Caffe Hrt
Heterogeneous Run Time version of Caffe. Added heterogeneous capabilities to the Caffe, uses heterogeneous computing infrastructure framework to speed up Deep Learning on Arm-based heterogeneous embedded platform. It also retains all the features of the original Caffe architecture which users deploy their applications seamlessly.
Stars: ✭ 271 (-20.06%)
Mutual labels: cnn, dnn
Python
This repository helps you understand python from the scratch.
Stars: ✭ 285 (-15.93%)
Mutual labels: jupyter-notebook, numpy
Pysynth
Several simple music synthesizers in Python 3. Input from ABC or MIDI files is also supported.
Stars: ✭ 279 (-17.7%)
Mutual labels: jupyter-notebook, numpy
Image Captioning
Image Captioning using InceptionV3 and beam search
Stars: ✭ 290 (-14.45%)
Mutual labels: jupyter-notebook, cnn
Natural Language Processing With Tensorflow
Natural Language Processing with TensorFlow, published by Packt
Stars: ✭ 222 (-34.51%)
Mutual labels: jupyter-notebook, cnn
Deform conv pytorch
PyTorch Implementation of Deformable Convolution
Stars: ✭ 217 (-35.99%)
Mutual labels: jupyter-notebook, cnn
Neural Network From Scratch
Ever wondered how to code your Neural Network using NumPy, with no frameworks involved?
Stars: ✭ 230 (-32.15%)
Mutual labels: jupyter-notebook, numpy
Tcdf
Temporal Causal Discovery Framework (PyTorch): discovering causal relationships between time series
Stars: ✭ 217 (-35.99%)
Mutual labels: jupyter-notebook, cnn
Style transfer
CNN image style transfer 🎨.
Stars: ✭ 210 (-38.05%)
Mutual labels: jupyter-notebook, cnn
Tianchi Medical Lungtumordetect
天池医疗AI大赛[第一季]：肺部结节智能诊断 UNet/VGG/Inception/ResNet/DenseNet
Stars: ✭ 314 (-7.37%)
Mutual labels: jupyter-notebook, vgg
Windrose
A Python Matplotlib, Numpy library to manage wind data, draw windrose (also known as a polar rose plot), draw probability density function and fit Weibull distribution
Stars: ✭ 208 (-38.64%)
Mutual labels: jupyter-notebook, numpy
3d Mri Brain Tumor Segmentation Using Autoencoder Regularization
Keras implementation of the paper "3D MRI brain tumor segmentation using autoencoder regularization" by Myronenko A. (https://arxiv.org/abs/1810.11654).
Stars: ✭ 209 (-38.35%)
Mutual labels: jupyter-notebook, cnn
numpy_neuron_network
仅使用numpy从头构建神经网络, 包括如下内容(持续更新中....)
-
网络中梯度反向传播公式推导
-
层：FC层,卷积层,池化层,Flatten
-
激活函数: ReLU、LeakyReLU、PReLU、ELU、SELU
-
损失函数：均方差、交叉熵
-
模型的保存、部署
-
案例学习：线性回归、图像分类
-
迁移学习、模型精调
-
进阶：RNN、LSTM、GRU、BN
[TOC]
运行工程
环境：python 3.6.x
依赖：numpy>=1.15.0、Cython、jupyter
a) 下载
git clone https://github.com/yizt/numpy_neuron_network
b) 编译nn/clayers.pyx
cd numpy_neuron_network
python setup.py build_ext -i
c) 启动工程,所有的notebook都可以直接运行
jupyter notebook --allow-root --ip 0.0.0.0
基础知识
0_2_1-卷积层的反向传播-单通道、无padding、步长1 、csdn地址
0_2_2-卷积层的反向传播-多通道、无padding、步长1 、csdn地址
0_2_3-卷积层的反向传播-多通道、无padding、步长不为1 、csdn地址
0_2_4-卷积层的反向传播-多通道、有padding、步长不为1 、csdn地址
0_2_5-池化层的反向传播-MaxPooling、AveragePooling、GlobalAveragePooling、GlobalMaxPooling 、csdn地址
0_3-激活函数的反向传播-ReLU、LeakyReLU、PReLU、ELU、SELU 、csdn地址
0_4-优化方法-SGD、AdaGrad、RMSProp、Adadelta、Adam 、csdn地址
DNN练习
1_1_2-全连接神经网络做mnist手写数字识别 、csdn地址
CNN练习
2_3-numpy-cnn-mnist手写数字识别 、csdn地址
2_4-对抗神经网络 、csdn地址
经典网络
进阶
5-1-RNN反向传播
5-2-LSTM反向传播
5-3-GRU反向传播
5-4-RNN、LSTM、GRU实现
5-5-案例-lstm连续文字识别
6-1-Batch Normalization反向传播
6-2-Batch Normalization实现
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].