pakddo/modu-tensorflow-v2

modulabs TensorFlow Tutorials for version 2.0

모두의연구소 풀잎스쿨 5기 Flow me to the moon 에서 활동하는 연구원들이 정리한 내용입니다. 아래의 참고자료를 기반으로 작성하고 있습니다. 앞으로 나올 TF version 2.0 에 맞게 작성하려고 노력중입니다.

• TensorFlow tutorials
  • Images
    • Image recognition, Neural style transfer, Image segmentation
  • Sequences
    • Recurrent neural network, Neural machine translation, Simple audio recognition
  • Generative models
    • DCGAN VAE, Image captioning
  • and so on
• CS 20: TensorFlow for Deep Learning Research (Stanford Lecture 2018)
  • Syllabus: http://web.stanford.edu/class/cs20si/syllabus.html
  • Github: https://github.com/chiphuyen/stanford-tensorflow-tutorials
• TensorFlow models

Getting Started

Prerequisites

• TensorFlow above 1.11.
  • We will use TensorFlow 2.0 later.
• Python 3.6
  • numpy, matplotlib, PIL
• Jupyter notebook
• OS X and Linux (Not validated on Windows but probably it would work)

Contents

TensorFlow overview

• 01.hello.tf.ipynb
• 02.tf.eager.ipynb
• 03.tensorboard.ipynb

TensorFlow basic operations

• 01-1.tf.constant.ipynb
• 01-2.tf.constant.eager.ipynb
• 02-1.tf.Variable.ipynb
• 02-2.tf.Variable.eager.ipynb
• 03.tf.placeholder.ipynb
• 04-1.normal.lazy.loading.ipynb
• 04-2.normal.lazy.loading.eager.ipynb

Regression

• 01-1.linear.regression.ipynb
• 01-2.linear.regression.eager.ipynb
• 02-1.logistic.regression.tf.placeholder.ipynb
• 02-2.logistic.regression.tf.data.ipynb
• 02-3.logistic.regression.tf.data.eager.ipynb

Images

Convolutional neural networks

• 01-1.cnn.mnist.tf.data.ipynb
• 01-2.cnn.mnist.tf.data.eager.ipynb
• 02-1.cnn.mnist.slim.ipynb
• 02-2.cnn.mnist.slim.eager.ipynb
• 03-1.cnn.mnist.keras.ipynb
• 03-2.cnn.mnist.keras.eager.ipynb

Style Transfer

Image segmentation

Sequences

Recurrent neural networks

Sequences to sequences

Audio segmentation

Generative models

Generative adversarial networks

Variational autoencoder

Image captioning