hylhero/finch

Lots of Machine Learning Models based on TensorFlow / PyTorch (Keep Updating)

finch is a project that contains the collection of my implementations of machine learning models and their tests based on modern frameworks (mainly TensorFlow and PyTorch).

I am actively responsible for this project, and there will be more models in the future ...

You are more than welcome to send an email to me in English / Chinese for any suggestion or question, I will try my best to answer as soon as I see it.

Installation

First get the project folder

git clone https://github.com/zhedongzheng/finch.git

Then run the test file that you are interested in

python xxxx_test.py

The main dependence is:

• TensorFlow 1.2
• PyTorch 0.12
• scikit-learn

Other Language Support

中文

Style of My Code

Most deep learning projects on GitHub usually have several scripts such as train.py, model.py, main.py and so on. However, personally I prefer integrating all the functionalities into a single model with fit() and predict() interfaces (scikit-learn API style), and then write separate test files for different incoming data. All the code below follow this style.

Contents

• Machine Learning
  • Linear Model
  • Support Vector Machine
  • Ensemble
  • Decomposition
• Deep Learning
  • Multilayer Perceptron
  • Convolutional Network
  • Recurrent Network
  • Autoencoder
  • Highway Network
  • Generative Adversarial Network
• Natural Language Processing
  • Preprocessing
  • Language Model
  • Text Classification
  • Text Generation
  • Sequence Labelling
  • Sequence to Sequence
• Computer Vision
  • OpenCV

Machine Learning

Linear Model

• TensorFlow   |   Linear Regression     Model     Test   |  

• TensorFlow   |   Logistic Regression     Model     Test   |  

• Java   |   Logistic Regression     Model     Test   |  

Support Vector Machine

• TensorFlow   |   Linear SVM Classifier     Model     Test   |  

• Java   |   Linear SVM Classifier     Model     Test   |  

• Libsvm   |   Non-linear SVM Classifier     Model     Test   |  

Ensemble

• NumPy   |   Bagging Classifier     Model     Test   |  

• NumPy   |   Adaboost Classifier     Pseudocode     Model     Test   |  

• NumPy   |   Random Forest Classifier     Model     Test   |  

Decomposition

• TensorFlow   |   Non-negative Matrix Factorization     Model     MovieLens Test   |  

Deep Learning

Multilayer Perceptron

• TensorFlow   |   MLP Classifier     Model     MNIST Test     CIFAR10 Test   |  

• PyTorch   |   MLP Classifier     Model     MNIST Test     CIFAR10 Test   |  

Convolutional Network

• TensorFlow   |   Conv2D Classifier     Model     MNIST Test     CIFAR10 Test   |  

• PyTorch   |   Conv2D Classifier     Model     MNIST Test     CIFAR10 Test   |  

Recurrent Network

• TensorFlow   |   LSTM Classifier     Model     MNIST Test     CIFAR10 Test   |  

• TensorFlow   |   LSTM Regressor     Model     Test     Preview   |  

• PyTorch   |   LSTM Classifier     Model     MNIST Test     CIFAR10 Test   |  

• PyTorch   |   GRU Regressor     Model     Test   |  

Autoencoder

• TensorFlow   |   MLP Autoencoder (weights-tied)     Model   |   MNIST Test   |  

• TensorFlow   |   Conv2D Autoencoder (weights-tied)     Model     MNIST Test     CIFAR10 Test   |  

Highway Network

• TensorFlow   |   MLP Highway Classifier     Model     MNIST Test   |  

• TensorFlow   |   Conv1D Highway Classifier     Model     IMDB Test   |  

Generative Adversarial Network

• TensorFlow   |   MLP GAN     Model     Test   |   MLP Conditional GAN     Model     Test   |  

• TensorFlow   |   DCGAN     MNIST     Model     Test     Result   |  

Natural Language Processing

Text Preprocessing

• Python   |   Text Cleaning

• Python   |   Word Indexing

Language Model

• Sklearn   |   Latent Semantic Analysis     Wiki     Model     Test   |  

• Python   |   Tri-Gram     Wiki     Model     Test   |  

• Sklearn   |   TF-IDF     Test     Result   |  

• TensorFlow   |   Word2vec (Skip-Gram)     Model     Text8 Test   |  

Text Classification

• Sklearn   |   TF-IDF + Logistic Regression     IMDB Test   |  

• TensorFlow   |   Conv1D     Model     IMDB Test   |   Concat Conv1D     Model     IMDB Test     Result   |  

• TensorFlow   |   LSTM     Model     IMDB Test   |  

• TensorFlow   |   Bi-directional LSTM + Attention     Model     IMDB Test     Result   |  

• TensorFlow   |   Conv1D-LSTM     Model     IMDB Test   |  

• PyTorch   |   Conv1D     Model     IMDB Test   |  

• PyTorch   |   LSTM     Model     IMDB Test   |  

• PyTorch   |   Conv1D-LSTM     Model     IMDB Test   |  

Text Generation

• Python   |   2nd order Markov     Model     Robert Frost Test   |  

• TensorFlow   |   Char-RNN     Karpathy's Idea     Model     Test   |  

• TensorFlow   |   CNN-RNN     Original Paper     Model     Test   |  

Sequence Labelling

• TensorFlow   |   LSTM     Model   |   POS Tagging Test   |   Chinese Segmentation Test   |  

• TensorFlow   |   Bi-directional LSTM     Model   |   POS Tagging Test   |   Chinese Segmentation Test   |  

• TensorFlow   |   Bi-directional LSTM + CRF     Model

  • POS Tagging Test   |   Result

  • Chinese Segmentation Test   |   Result

• PyTorch   |   LSTM     Model   |   POS Tagging Test   |   Chinese Segmentation Test   |  

• PyTorch   |   Bi-directional LSTM     Model   |   POS Tagging Test   |   Chinese Segmentation Test   |  

Sequence to Sequence

• TensorFlow   |   Dynamic Seq2Seq     Model

  • Sorting Test

• TensorFlow   |   Dynamic Seq2Seq + Bi-directional Encoder     Model

  • Sorting Test

• TensorFlow   |   Dynamic Seq2Seq + Attention Mechanism     Model    

  • Sorting Test     Result

Computer Vision

OpenCV

• OP   |   Resize

• OP   |   Rotations

• Segmentation   |   Contours

• Segmentation   |   Sorting Contours

• Detection   |   Face & Eye Detection Using Cascade Classifier

• Detection   |   Walker & Car Detection Using Cascade Classifier