Merge pull request #112801 from minthigpen/interpret-doc-typofix

Interpret doc typofix

Author: PRMerger12

articles/machine-learning/how-to-machine-learning-interpretability-aml.md

@@ -233,10 +233,10 @@ The following example shows how you can use the `ExplanationClient` class to ena
 * Use the `ExplanationClient` in the remote run to upload the interpretability context.
 * Download the context later in a local environment.
 
-1. Install `azureml-interpret` and `azureml-interpret-contrib` packages.
+1. Install `azureml-interpret` and `azureml-contrib-interpret` packages.
     ```bash
     pip install azureml-interpret
-    pip install azureml-interpret-contrib
+    pip install azureml-contrib-interpret
     ```
 1. Create a training script in a local Jupyter notebook. For example, `train_explain.py`.
 

articles/machine-learning/how-to-machine-learning-interpretability.md

@@ -71,17 +71,17 @@ Learn about supported interpretability techniques, supported machine learning mo
 
 |Interpretability Technique|Description|Type|
 |--|--|--------------------|
-|1. SHAP Tree Explainer| [SHAP](https://github.com/slundberg/shap)'s tree explainer, which focuses on polynomial time fast SHAP value estimation algorithm specific to **trees and ensembles of trees**.|Model-specific|
-|2. SHAP Deep Explainer| Based on the explanation from [SHAP](https://github.com/slundberg/shap), Deep Explainer "is a high-speed approximation algorithm for SHAP values in deep learning models that builds on a connection with DeepLIFT described in the [SHAP NIPS paper](https://papers.nips.cc/paper/7062-a-unified-approach-to-interpreting-model-predictions). **TensorFlow** models and **Keras** models using the TensorFlow backend are supported (there is also preliminary support for PyTorch)".|Model-specific|
-|3. SHAP Linear Explainer| [SHAP](https://github.com/slundberg/shap)'s Linear explainer computes SHAP values for a **linear model**, optionally accounting for inter-feature correlations.|Model-specific|
-|4. SHAP Kernel Explainer| [SHAP](https://github.com/slundberg/shap)'s Kernel explainer uses a specially weighted local linear regression to estimate SHAP values for **any model**.|Model-agnostic|
-|5. Mimic Explainer (Global Surrogate)| Mimic explainer is based on the idea of training [global surrogate models](https://christophm.github.io/interpretable-ml-book/global.html) to mimic blackbox models. A global surrogate model is an intrinsically interpretable model that is trained to approximate the predictions of **any black box model** as accurately as possible. Data scientists can interpret the surrogate model to draw conclusions about the black box model. You can use one of the following interpretable models as your surrogate model: LightGBM (LGBMExplainableModel), Linear Regression (LinearExplainableModel), Stochastic Gradient Descent explainable model (SGDExplainableModel), and Decision Tree (DecisionTreeExplainableModel).|Model-agnostic|
-|6. Permutation Feature Importance Explainer (PFI)| Permutation Feature Importance is a technique used to explain classification and regression models that is inspired by [Breiman's Random Forests paper](https://www.stat.berkeley.edu/~breiman/randomforest2001.pdf) (see section 10). At a high level, the way it works is by randomly shuffling data one feature at a time for the entire dataset and calculating how much the performance metric of interest changes. The larger the change, the more important that feature is. PFI can explain the overall behavior of **any underlying model** but does not explain individual predictions. |Model-agnostic|
+|SHAP Tree Explainer| [SHAP](https://github.com/slundberg/shap)'s tree explainer, which focuses on polynomial time fast SHAP value estimation algorithm specific to **trees and ensembles of trees**.|Model-specific|
+|SHAP Deep Explainer| Based on the explanation from SHAP, Deep Explainer "is a high-speed approximation algorithm for SHAP values in deep learning models that builds on a connection with DeepLIFT described in the [SHAP NIPS paper](https://papers.nips.cc/paper/7062-a-unified-approach-to-interpreting-model-predictions). **TensorFlow** models and **Keras** models using the TensorFlow backend are supported (there is also preliminary support for PyTorch)".|Model-specific|
+|SHAP Linear Explainer| SHAP's Linear explainer computes SHAP values for a **linear model**, optionally accounting for inter-feature correlations.|Model-specific|
+|SHAP Kernel Explainer| SHAP's Kernel explainer uses a specially weighted local linear regression to estimate SHAP values for **any model**.|Model-agnostic|
+|Mimic Explainer (Global Surrogate)| Mimic explainer is based on the idea of training [global surrogate models](https://christophm.github.io/interpretable-ml-book/global.html) to mimic blackbox models. A global surrogate model is an intrinsically interpretable model that is trained to approximate the predictions of **any black box model** as accurately as possible. Data scientists can interpret the surrogate model to draw conclusions about the black box model. You can use one of the following interpretable models as your surrogate model: LightGBM (LGBMExplainableModel), Linear Regression (LinearExplainableModel), Stochastic Gradient Descent explainable model (SGDExplainableModel), and Decision Tree (DecisionTreeExplainableModel).|Model-agnostic|
+|Permutation Feature Importance Explainer (PFI)| Permutation Feature Importance is a technique used to explain classification and regression models that is inspired by [Breiman's Random Forests paper](https://www.stat.berkeley.edu/~breiman/randomforest2001.pdf) (see section 10). At a high level, the way it works is by randomly shuffling data one feature at a time for the entire dataset and calculating how much the performance metric of interest changes. The larger the change, the more important that feature is. PFI can explain the overall behavior of **any underlying model** but does not explain individual predictions. |Model-agnostic|
 
 
 
 
-Besides the interpretability techniques described above, we support another [SHAP-based explainer](https://github.com/slundberg/shap), called `TabularExplainer`. Depending on the model, `TabularExplainer` uses one of the supported SHAP explainers:
+Besides the interpretability techniques described above, we support another SHAP-based explainer, called `TabularExplainer`. Depending on the model, `TabularExplainer` uses one of the supported SHAP explainers:
 
 * TreeExplainer for all tree-based models
 * DeepExplainer for DNN models
@@ -117,4 +117,6 @@ You can run explanation remotely on Azure Machine Learning Compute and log the e
 
 ## Next steps
 
-See the [how-to](how-to-machine-learning-interpretability-aml.md) for enabling interpretability for models training both locally and on Azure Machine Learning remote compute resources. See the [sample notebooks](https://github.com/Azure/MachineLearningNotebooks/tree/master/how-to-use-azureml/explain-model) for additional scenarios.
+- See the [how-to](how-to-machine-learning-interpretability-aml.md) for enabling interpretability for models training both locally and on Azure Machine Learning remote compute resources. 
+- See the [sample notebooks](https://github.com/Azure/MachineLearningNotebooks/tree/master/how-to-use-azureml/explain-model) for additional scenarios. 
+- If you're interested in interpretability for text scenarios, see [Interpret-text](https://github.com/interpretml/interpret-text), a related open source repo to [Interpret-Community](https://github.com/interpretml/interpret-community/), for interpretability techniques for NLP. `azureml.interpret` package does not currently support these techniques but you can get started with an [example notebook on text classification](https://github.com/interpretml/interpret-text/blob/master/notebooks/text_classification/text_classification_classical_text_explainer.ipynb).