Skip to content

Latest commit

 

History

History

GSP710_Using-the-What-If-Tool-with-Image-Recognition-Models

Folders and files

NameName
Last commit message
Last commit date

parent directory

..

README.md

README.md

 
 

README.md

GSP710 —— Using the What-If Tool with Image Recognition Models

Table of Contents （🔎 Click to expand／collapse）

Overview

In this lab, you explore the use of the What-If Tool (WIT) for image recognition models. Your task is to predict whether a person is smiling. The lab provides a CNN (Convolutional Neural Network) that is trained on a subset of CelebA dataset and visualizes the results on a separate test subset.

Objectives:

  • Launch an AI Platform Notebook.
  • Download the pre-trained Keras model.
  • Define helper functions for dataset conversion from csv to tf.
  • Load the csv file into pandas dataframe and process it for WIT.
  • Load the Keras models.
  • Define the custom predict function for WIT.

What-If Tool

We can use the What-If Tool (WIT) within notebook environments to inspect AI Platform Prediction models through an interactive dashboard.

  • An extension in Jupyter, Colaboratory, and Cloud AI Platform notebooks.
  • Integrates with TensorBoard, Jupyter notebooks, Colab notebooks, and JupyterHub.
  • Pre-installed on Notebooks TensorFlow instances.
  • Used to analyze classification or regression models on datapoints as inputs directly from within the notebook.

References