DeepLearning Notebooks

These are deep learning examples implemented by TensorFlow, Python with Numpy.

Prerequisites

All python code are base on python3 and use jupyter notebook.

• TensorFlow 1.0
• Scikit Learn
• matplotlib, Seaborn
• Numpy
• Pandas

DataSet

1. Boston Housing DataSet
2. Mnist
3. CIFAR-10
4. iris
5. polarity dataset v2.0i (Movie Review)

Contents

Python, TensorFlow is minimal version.
Exercise is used model that mimicking scikit-learn's interface (fit, predict, etc...) and extended version from minimal.

1. Linear Regression
   Python | TensorFlow | Exercise - Boston Housing(TensorFlow)
2. Logistic Regression
   TensorFlow | Exercies - Iris(Python)
3. Neural Network
   • Two Layer Network: Python) | AND, OR, XOR(Python)-2_Layer
   • N Layer Network: Exercise - AND, OR, XOR(Python)-N_Layer , TensorFlow
4. Convolutional Neural Network
   TensorFlow(AlexNet)
5. Recurrent Neural Network
   TensorFlow
6. Word2Vec
   TensorFlow
7. CNN for Sentence Classification
   PDF | TensorFlow
8. Char-RNN (a character-level language model to generate character sequences)
   TensorFlow(Obama-RNN)
9. Seq2Seq
   TensorFlow
10. Adversarial Neural Cryptography
    TensorFlow

Reference

• CS231n Convolutional Neural Networks for Visual Recognition
• CS224d: Deep Learning for Natural Language Processing
• TensorFlow Tutorial
• WILDML
• R2RT