Update README.md

02_data_science_toolkit/03_explaining_image_classifier_predictions/README.md

@@ -32,7 +32,11 @@ cd 01_coco_res_net
 sh get_coco_dataset_sample.sh
 ```
 
-## ELI5 example
+## ELI5
+
+[Full code example][4]
+
+The first library we will look into is Eli5 - it's a simple but reliable tool designed to visualize, inspect and debug ML models. The library allows, among other things, to interpret the predictions of Image Classifiers written in Keras. To do it, Eli5 leverages Gradient-weighted Class Activation Mapping (Grad-CAM) algorithm. It is worth noting that this is not a general approach and it applies only to CNN solutions.
 
 <p align="center"> 
     <img width="700" src="./01_coco_res_net/viz/coco_resnet34_eli5.png" alt="Eli5">
@@ -42,6 +46,38 @@ sh get_coco_dataset_sample.sh
     <b>Figure 1.</b> Explanations provided by ELI5
 </p>
 
+## LIME
+
+[Full code example][5]
+
+[Why Should I Trust You?: Explaining the Predictions of Any Classifier][7] is an article that underlies the entire branch of research aimed at explaining ML models. The ideas included in this paper became the foundation for the most popular of the interpretation libraries - Local interpretable model-agnostic explanations (LIME). This algorithm is completely different from Grad-CAM - tries to understand the model by perturbating the input data and understanding how these changes affect the predictions.
+
+<p align="center"> 
+    <img width="700" src="./01_coco_res_net/viz/coco_resnet34_lime.png" alt="LIME">
+</p>
+
+<p align="center"> 
+    <b>Figure 1.</b> Explanations provided by LIME
+</p>
+
+## SHAP
+
+[Full code example][6]
+
+SHapley Additive exPlanations (SHAP) and LIME are quite similar - both are additive and model-agnostic methods of explaining individual predictions. SHAP however, aims to explain model prediction for given input by computing the contribution of each feature to this prediction. To achieve this goal, SHAP uses Shapley Values, which originally comes from game theory.
+
+<p align="center"> 
+    <img width="700" src="./01_coco_res_net/viz/coco_resnet34_shap.png" alt="SHAP">
+</p>
+
+<p align="center"> 
+    <b>Figure 1.</b> Explanations provided by SHAP
+</p>
+
 [1]: https://github.com/TeamHG-Memex/eli5
 [2]: https://github.com/marcotcr/lime
 [3]: https://github.com/slundberg/shap
+[4]: "./01_coco_res_net/coco-resnet-keras-resnet-34-eli5.ipynb"
+[5]: "./01_coco_res_net/coco-resnet-keras-resnet-34-lime.ipynb"
+[6]: "./01_coco_res_net/coco-resnet-keras-resnet-34-shap.ipynb"
+[7]: https://arxiv.org/abs/1602.04938