/learn-tensorflow

TensorFlow oficial documentation tutorials.

Primary LanguagePythonMIT LicenseMIT

TensorFlow Tutorials - Google Oficial Documentation

image

Description

In this repo I re-implement the TensorFlow tutorials in order to master the concepts and tools available to implement deep neural networks to solve all sort of machine learning problems. If you want to check the oficial website click in this link.

Summary

Following the official documentation, I split the content in topics and sub topics such that:

  • Beginner
    • ML basics with Keras
      1. Basic image classification
      2. Basic text classification
      3. Text classification with TF Hub
      4. Regression
      5. Overfit and Underfit
      6. Save and load
      7. Tune hyperparameters with the Keras Tuner
      8. Other examples on keras.io
    • Load and preprocess data
      1. Images
      2. CSV
      3. NumPy
      4. pandas.DataFrame
      5. Text
      6. Unicode
      7. TF.Text
      8. Subword tokenization
      9. TFRecord and tf.Example
      10. Addtional examples with tf.io
  • Advanced
    • Customization
      1. Tensor and operations
      2. Custom layers
      3. Custom training: walkthrough
    • Distributed training
      1. Distributed training with Keras
      2. Custom training loops
      3. Multi-worker training with Keras
      4. Parameter server trainer
      5. Save and load
      6. Distributed input
    • Images
      1. Convolutional Neural Network
      2. Image Classification
      3. Transfer learning and fine-tuning
      4. Transfer learning with TF Hub
      5. Data augmentation
      6. Image segmentation
      7. Object detection with TF Hub
    • Text
      1. Word embedding
      2. Word2Vec
      3. Text classification with RNN
      4. Classify text with BERT
      5. Solve GLUE tasks using BERT on TPU
      • Generation
        1. Text generation with an RNN
        2. Neural machine translation with attention
        3. Image captionign
        4. Transformer model for language understanding
    • Audio
      1. Simple audio recognition
      2. Transfer learning for audio recognition
    • Structured Data
      1. Classify structured data with feature columns
      2. Classify structured data with preprocessing layers
      3. Classification on imbalanced data
      4. Time series forecasting
      5. Recommenders
    • Generative
      1. Neural style transfer
      2. DeepDream
      3. DCGAN
      4. Pix2Pix
      5. CycleGAN
      6. Adversarial FGSM
      7. Intro to Autoencoders
      8. Variational Autoencoders
    • Interpretability
      1. Integrated gradients
    • Reinforcement learning
      1. Actor-Critic method
      2. TensorFlow agents
    • tf.Estimator
      1. Premade estimator
      2. Linear model
      3. Bossted trees
      4. Boosted trees model understanding
      5. Keras model estimator
      6. Multi-Worker training with Estimator

Requirements

  • Python
  • NumPy
  • Pandas
  • Matplotlib
  • Seaborn
  • PyYAML
  • H5Py
  • Tensorflow
  • Tensorflow Docs
  • Tensorflow Hub
  • TensorFlow Text