Pro Deep Learning with TensorFlow by Santanu Pattanayak

Author:Santanu Pattanayak
Language: eng
Format: epub
Publisher: Apress, Berkeley, CA

Google’s TensorFlow provides a pre-trained model that is trained on the ImageNet data. It can be used for transfer learning. We use the pre-trained model from Google and retrain it on a set of cat versus dog images extracted from https://www.kaggle.com/c/dogs-vs-cats/data . The train.zip dataset contains 25,000 images, with 12,500 images each for cats and dogs.

The pre-trained model can be found in the TensorFlow GitHub Examples folder. Listing 3-8 shows the steps one needs to follow in order to execute and use the model for transfer-learning purposes. Clone the TensorFlow GitHub repository, since the model sits within it in the Examples folder. Once done, we are all set to enter the cloned TensorFlow folder and execute the commands in Listing 3-8.

#Step1 - Download the following dataset and un-tar it since it would be used for building the retrainer for transfer learning

Listing 3-8.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.