tutorial.md

Tutorial

This tutorial provides instructions (with examples), basic and end-to-end examples to help user to learn optimize models by Intel® Neural Compressor.

Instructions with Examples

Usage Example

Helloworld Examples

1. Built-in dataloader and metric example: see tf_example1 for more details.
2. TensorFlow checkpoint: see tf_example4 for more details.
3. Enable benchmark for performance and accuracy measurement: see tf_example5 for more details.
4. TensorFlow slim model: see tf_example3 for more details.

Tutorial Examples

These examples are easy to run without preparing dataset. The script will download public dataset automatically or create fake dataset.

1. Compare existed FP32 & INT8 ResNet50 model directly: Performance of FP32 Vs. INT8 ResNet50 Model.
2. An End-To-End pipeline to build up a CNN model by Pytorch to recognize fashion image and speed up AI model by Intel® Neural Compressor: Intel® Neural Compressor Sample for PyTorch*
3. An End-To-End pipeline to build up a CNN model by Tensorflow to recognize handwriting number and speed up AI model by Intel® Neural Compressor: Intel® Neural Compressor Sample for TensorFlow*