tf_image_classification

Transfer Learning for Image Classification with TF Hub

This notebook uses transfer learning with multiple TF Hub image classifiers, TF datasets, and custom image datasets.

The notebook performs the following steps:

1. Import dependencies and setup parameters
2. Prepare the dataset
3. Predict using the original model
4. Transfer Learning
5. Evaluate the model
6. Export the saved model

Running the notebook

To run the notebook, follow the instructions to setup the TensorFlow notebook environment.

References

Dataset citations

@inproceedings{bossard14,
  title = {Food-101 -- Mining Discriminative Components with Random Forests},
  author = {Bossard, Lukas and Guillaumin, Matthieu and Van Gool, Luc},
  booktitle = {European Conference on Computer Vision},
  year = {2014}
}

@ONLINE {tfflowers,
author = "The TensorFlow Team",
title = "Flowers",
month = "jan",
year = "2019",
url = "http://download.tensorflow.org/example_images/flower_photos.tgz" }