BiT Image Classification Quantization NNCF notebook (#1571)

* BiT Model Optimization (FP32) and Quantization (INT8) pipeline with OpenVINO and NNCF

* Update

* BiT Image Classification OpenVINO model quantization pipeline with NNCF

* Rename big transfer notebook directory

* Rename big transfer directory

* Readme.md

* Arrange code sections and add titles

* Update readme.md

* Update log messages

* BiT image classification model quantization with NNCF

* Remove unused imports

* Fix build review check

* Fix build review check

* Fix build review check

* Fix build review check

* Fix build review check

* Update README.md

Add 127-big-transfer-quantization notebook

* Fix build review check

* Address review comments

* Fix segmentation fault with nncf>=2.7.0 jra-Ccvs129152

* Add BiT to spell check script

* docker_teron test fix

* Update table of contents

---------

Co-authored-by: Ekaterina Aidova <ekaterina.aidova@intel.com>

Author: spradeep28

.ci/spellcheck/.pyspelling.wordlist.txt

@@ -53,6 +53,7 @@ Bicubic
 bicubic
 bilinear
 biometrics
+BiT
 blockwise
 boolean
 CentOS

README.md

@@ -142,6 +142,7 @@ Tutorials that explain how to optimize and quantize models with OpenVINO tools.
 | [124-hugging-face-hub](notebooks/124-hugging-face-hub/)<br>[![Binder](https://mybinder.org/badge_logo.svg)](https://mybinder.org/v2/gh/openvinotoolkit/openvino_notebooks/HEAD?filepath=notebooks%2F124-hugging-face-hub%2F124-hugging-face-hub.ipynb)<br>[![Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/openvinotoolkit/openvino_notebooks/blob/main/notebooks/124-hugging-face-hub/124-hugging-face-hub.ipynb) | Load models from Hugging Face Model Hub with OpenVINO™ |
 | [125-torchvision-zoo-to-openvino](notebooks/125-torchvision-zoo-to-openvino/)<br>Classification<br>[![Binder](https://mybinder.org/badge_logo.svg)](https://mybinder.org/v2/gh/openvinotoolkit/openvino_notebooks/HEAD?filepath=notebooks%2F125-torchvision-zoo-to-openvino%2F125-convnext-classification.ipynb)<br>[![Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/openvinotoolkit/openvino_notebooks/blob/main/notebooks/125-torchvision-zoo-to-openvino/125-convnext-classification.ipynb)<br>Semantic Segmentation<br>[![Binder](https://mybinder.org/badge_logo.svg)](https://mybinder.org/v2/gh/openvinotoolkit/openvino_notebooks/HEAD?filepath=notebooks%2F125-torchvision-zoo-to-openvino%2F125-lraspp-segmentation.ipynb)<br>[![Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/openvinotoolkit/openvino_notebooks/blob/main/notebooks/125-torchvision-zoo-to-openvino/125-lraspp-segmentation.ipynb)| Convert torchvision classification and semantic segmentation models to OpenVINO IR |
 | [126-tensorflow-hub](notebooks/126-tensorflow-hub/)<br>[![Binder](https://mybinder.org/badge_logo.svg)](https://mybinder.org/v2/gh/openvinotoolkit/openvino_notebooks/HEAD?filepath=notebooks%2F126-tensorflow-hub%2F126-tensorflow-hub.ipynb)<br>[![Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/openvinotoolkit/openvino_notebooks/blob/main/notebooks/126-tensorflow-hub/126-tensorflow-hub.ipynb) | Convert TensorFlow Hub models to OpenVINO IR |
+| [127-big-transfer-quantization](notebooks/127-big-transfer-quantization/)| BiT Image Classification OpenVINO IR model Quantization with NNCF |
 
 <div id='-model-demos'></div>
 

notebooks/127-big-transfer-quantization/127-tensorflow-bit-image-classification-nncf-quantization.ipynb

@@ -0,0 +1,21 @@
+# Big Transfer Image Classification Model Quantization with NNCF in OpenVINO™
+
+This tutorial demonstrates how to apply 'INT8' quantization to the [Big Transfer](https://tfhub.dev/google/bit/m-r50x1) Image Classification model. Here we demonstrate fine-tuning the model, OpenVINO optimization and followed by INT8 quantization processing with [NNCF](https://github.com/openvinotoolkit/nncf/).
+
+## Notebook Contents
+
+This tutorial consists of the following steps:
+- Prepare Dataset.
+- Plotting data samples.
+- Model fine-tuning.
+- Perform model optimization (IR) step.
+- Compute model accuracy of the TF model.
+- Compute model accuracy of the optimized model.
+- Run nncf.Quantize for getting an Optimized model.
+- Compute model accuracy of the quantized model.
+- Compare model accuracy of the optimized and quantized models.
+- Compare inference results on one picture 
+
+## Installation Instructions
+
+If you have not installed all required dependencies, follow the [Installation Guide](https://github.com/openvinotoolkit/openvino_notebooks/blob/main/README.md#-installation-guide)