Overview
This course teaches how to create an input pipeline for text data using TextLineDataset in TensorFlow. The learning outcomes include loading text data from files, filtering datasets, creating vocabulary, numericalizing with TokenTextEncoder, applying map on datasets, and building a simple model. The course covers handling text data structured in various ways, such as split over multiple files or in translation datasets. The intended audience for this course is individuals interested in working with text data in machine learning and TensorFlow.
Syllabus
- Introduction and Dataset Overview
- Load using TextLineDataset
- Filtering Dataset
- Creating Vocabulary
- Numericalizing with TokenTextEncoder
- Applying map on datasets
- Simple Model
- Dataset in Several Files
- Sketch Load Translation Dataset
- Ending
Taught by
Aladdin Persson