# Alibi Explain ## alibi explain - [Introduction](https://docs.seldon.ai/alibi-explain/overview/high_level.md) - [Getting Started](https://docs.seldon.ai/alibi-explain/overview/getting_started.md) - [Algorithm Overview](https://docs.seldon.ai/alibi-explain/overview/algorithms.md) - [White-box and black-box models](https://docs.seldon.ai/alibi-explain/overview/white_box_black_box.md) - [Saving and loading](https://docs.seldon.ai/alibi-explain/overview/saving.md) - [Frequently Asked Questions](https://docs.seldon.ai/alibi-explain/overview/faq.md) - [Methods](https://docs.seldon.ai/alibi-explain/explanations/methods.md) - [ALE](https://docs.seldon.ai/alibi-explain/explanations/methods/ale.md) - [Anchors](https://docs.seldon.ai/alibi-explain/explanations/methods/anchors.md) - [CEM](https://docs.seldon.ai/alibi-explain/explanations/methods/cem.md) - [CF](https://docs.seldon.ai/alibi-explain/explanations/methods/cf.md) - [CFProto](https://docs.seldon.ai/alibi-explain/explanations/methods/cfproto.md) - [CFRL](https://docs.seldon.ai/alibi-explain/explanations/methods/cfrl.md) - [IntegratedGradients](https://docs.seldon.ai/alibi-explain/explanations/methods/integratedgradients.md) - [KernelSHAP](https://docs.seldon.ai/alibi-explain/explanations/methods/kernelshap.md) - [LinearityMeasure](https://docs.seldon.ai/alibi-explain/explanations/methods/linearitymeasure.md) - [PartialDependence](https://docs.seldon.ai/alibi-explain/explanations/methods/partialdependence.md) - [PartialDependenceVariance](https://docs.seldon.ai/alibi-explain/explanations/methods/partialdependencevariance.md) - [PermutationImportance](https://docs.seldon.ai/alibi-explain/explanations/methods/permutationimportance.md) - [ProtoSelect](https://docs.seldon.ai/alibi-explain/explanations/methods/protoselect.md) - [Similarity](https://docs.seldon.ai/alibi-explain/explanations/methods/similarity.md) - [TreeSHAP](https://docs.seldon.ai/alibi-explain/explanations/methods/treeshap.md) - [TrustScores](https://docs.seldon.ai/alibi-explain/explanations/methods/trustscores.md) - [Examples](https://docs.seldon.ai/alibi-explain/explanations/examples.md) - [Alibi Overview Examples](https://docs.seldon.ai/alibi-explain/explanations/examples/overview.md) - [Accumulated Local Effets](https://docs.seldon.ai/alibi-explain/explanations/examples/accumulated-local-effets.md) - [Accumulated Local Effects for classifying flowers](https://docs.seldon.ai/alibi-explain/explanations/examples/accumulated-local-effets/ale_classification.md) - [Accumulated Local Effects for predicting house prices](https://docs.seldon.ai/alibi-explain/explanations/examples/accumulated-local-effets/ale_regression_california.md) - [Anchors](https://docs.seldon.ai/alibi-explain/explanations/examples/anchors.md) - [Anchor explanations for fashion MNIST](https://docs.seldon.ai/alibi-explain/explanations/examples/anchors/anchor_image_fashion_mnist.md) - [Anchor explanations for ImageNet](https://docs.seldon.ai/alibi-explain/explanations/examples/anchors/anchor_image_imagenet.md) - [Anchor explanations for income prediction](https://docs.seldon.ai/alibi-explain/explanations/examples/anchors/anchor_tabular_adult.md) - [Anchor explanations on the Iris dataset](https://docs.seldon.ai/alibi-explain/explanations/examples/anchors/anchor_tabular_iris.md) - [Anchor explanations for movie sentiment](https://docs.seldon.ai/alibi-explain/explanations/examples/anchors/anchor_text_movie.md) - [Contrastive Explanation Method](https://docs.seldon.ai/alibi-explain/explanations/examples/contrastive-explanation-method.md) - [Contrastive Explanations Method (CEM) applied to Iris dataset](https://docs.seldon.ai/alibi-explain/explanations/examples/contrastive-explanation-method/cem_iris.md) - [Contrastive Explanations Method (CEM) applied to MNIST](https://docs.seldon.ai/alibi-explain/explanations/examples/contrastive-explanation-method/cem_mnist.md) - [Counterfactual Instances on MNIST](https://docs.seldon.ai/alibi-explain/explanations/examples/cf_mnist.md) - [Counterfactuals Guided by Prototypes](https://docs.seldon.ai/alibi-explain/explanations/examples/counterfactuals-guided-by-prototypes.md) - [Counterfactual explanations with one-hot encoded categorical variables](https://docs.seldon.ai/alibi-explain/explanations/examples/counterfactuals-guided-by-prototypes/cfproto_cat_adult_ohe.md) - [Counterfactual explanations with ordinally encoded categorical variables](https://docs.seldon.ai/alibi-explain/explanations/examples/counterfactuals-guided-by-prototypes/cfproto_cat_adult_ord.md) - [Counterfactuals guided by prototypes on California housing dataset](https://docs.seldon.ai/alibi-explain/explanations/examples/counterfactuals-guided-by-prototypes/cfproto_housing.md) - [Counterfactuals guided by prototypes on MNIST](https://docs.seldon.ai/alibi-explain/explanations/examples/counterfactuals-guided-by-prototypes/cfproto_mnist.md) - [Counterfactuals with Reinforcement Learning](https://docs.seldon.ai/alibi-explain/explanations/examples/counterfactuals-with-reinforcement-learning.md) - [Counterfactual with Reinforcement Learning (CFRL) on Adult Census](https://docs.seldon.ai/alibi-explain/explanations/examples/counterfactuals-with-reinforcement-learning/cfrl_adult.md) - [Counterfactual with Reinforcement Learning (CFRL) on MNIST](https://docs.seldon.ai/alibi-explain/explanations/examples/counterfactuals-with-reinforcement-learning/cfrl_mnist.md) - [Integrated Gradients](https://docs.seldon.ai/alibi-explain/explanations/examples/integrated-gradients.md) - [Integrated gradients for a ResNet model trained on Imagenet dataset](https://docs.seldon.ai/alibi-explain/explanations/examples/integrated-gradients/integrated_gradients_imagenet.md) - [Integrated gradients for text classification on the IMDB dataset](https://docs.seldon.ai/alibi-explain/explanations/examples/integrated-gradients/integrated_gradients_imdb.md) - [Integrated gradients for MNIST](https://docs.seldon.ai/alibi-explain/explanations/examples/integrated-gradients/integrated_gradients_mnist.md) - [Integrated gradients for transformers models](https://docs.seldon.ai/alibi-explain/explanations/examples/integrated-gradients/integrated_gradients_transformers.md) - [Kernel SHAP](https://docs.seldon.ai/alibi-explain/explanations/examples/kernel-shap.md) - [Distributed KernelSHAP](https://docs.seldon.ai/alibi-explain/explanations/examples/kernel-shap/distributed_kernel_shap_adult_lr.md) - [KernelSHAP: combining preprocessor and predictor](https://docs.seldon.ai/alibi-explain/explanations/examples/kernel-shap/kernel_shap_adult_categorical_preproc.md) - [Handling categorical variables with KernelSHAP](https://docs.seldon.ai/alibi-explain/explanations/examples/kernel-shap/kernel_shap_adult_lr.md) - [Kernel SHAP explanation for SVM models](https://docs.seldon.ai/alibi-explain/explanations/examples/kernel-shap/kernel_shap_wine_intro.md) - [Kernel SHAP explanation for multinomial logistic regression models](https://docs.seldon.ai/alibi-explain/explanations/examples/kernel-shap/kernel_shap_wine_lr.md) - [Partial Dependence](https://docs.seldon.ai/alibi-explain/explanations/examples/partial-dependence.md) - [Partial Dependence and Individual Conditional Expectation for predicting bike renting](https://docs.seldon.ai/alibi-explain/explanations/examples/partial-dependence/pdp_regression_bike.md) - [Partial Dependence Variance](https://docs.seldon.ai/alibi-explain/explanations/examples/partial-dependence-variance.md) - [Feature importance and feature interaction based on partial dependece variance](https://docs.seldon.ai/alibi-explain/explanations/examples/partial-dependence-variance/pd_variance_regression_friedman.md) - [Permutation Importance](https://docs.seldon.ai/alibi-explain/explanations/examples/permutation-importance.md) - [Permutation Feature Importance on “Who’s Going to Leave Next?”](https://docs.seldon.ai/alibi-explain/explanations/examples/permutation-importance/permutation_importance_classification_leave.md) - [Similarity explanations](https://docs.seldon.ai/alibi-explain/explanations/examples/similarity-explanations.md) - [Similarity explanations for 20 newsgroups dataset](https://docs.seldon.ai/alibi-explain/explanations/examples/similarity-explanations/similarity_explanations_20ng.md) - [Similarity explanations for ImageNet](https://docs.seldon.ai/alibi-explain/explanations/examples/similarity-explanations/similarity_explanations_imagenet.md) - [Similarity explanations for MNIST](https://docs.seldon.ai/alibi-explain/explanations/examples/similarity-explanations/similarity_explanations_mnist.md) - [Tree SHAP](https://docs.seldon.ai/alibi-explain/explanations/examples/tree-shap.md) - [Explaining Tree Models with Interventional Feature Perturbation Tree SHAP](https://docs.seldon.ai/alibi-explain/explanations/examples/tree-shap/interventional_tree_shap_adult_xgb.md) - [Explaining Tree Models with Path-Dependent Feature Perturbation Tree SHAP](https://docs.seldon.ai/alibi-explain/explanations/examples/tree-shap/path_dependent_tree_shap_adult_xgb.md) - [Methods](https://docs.seldon.ai/alibi-explain/model-confidence/methods.md) - [Examples](https://docs.seldon.ai/alibi-explain/model-confidence/examples.md) - [Measuring the linearity of machine learning models](https://docs.seldon.ai/alibi-explain/model-confidence/examples/measuring-the-linearity-of-machine-learning-models.md) - [Linearity measure applied to fashion MNIST](https://docs.seldon.ai/alibi-explain/model-confidence/examples/measuring-the-linearity-of-machine-learning-models/linearity_measure_fashion_mnist.md) - [Linearity measure applied to Iris](https://docs.seldon.ai/alibi-explain/model-confidence/examples/measuring-the-linearity-of-machine-learning-models/linearity_measure_iris.md) - [Trust Scores](https://docs.seldon.ai/alibi-explain/model-confidence/examples/trust-scores.md) - [Trust Scores applied to Iris](https://docs.seldon.ai/alibi-explain/model-confidence/examples/trust-scores/trustscore_iris.md) - [Trust Scores applied to MNIST](https://docs.seldon.ai/alibi-explain/model-confidence/examples/trust-scores/trustscore_mnist.md) - [Methods](https://docs.seldon.ai/alibi-explain/prototypes/methods.md) - [Examples](https://docs.seldon.ai/alibi-explain/prototypes/examples.md) - [ProtoSelect on Adult Census and CIFAR10](https://docs.seldon.ai/alibi-explain/prototypes/examples/protoselect_adult_cifar10.md) - [alibi.api](https://docs.seldon.ai/alibi-explain/api-reference/api.md) - [alibi.api.defaults](https://docs.seldon.ai/alibi-explain/api-reference/api/defaults.md) - [alibi.api.interfaces](https://docs.seldon.ai/alibi-explain/api-reference/api/interfaces.md) - [alibi.confidence](https://docs.seldon.ai/alibi-explain/api-reference/confidence.md) - [alibi.confidence.model\_linearity](https://docs.seldon.ai/alibi-explain/api-reference/confidence/model_linearity.md) - [alibi.confidence.trustscore](https://docs.seldon.ai/alibi-explain/api-reference/confidence/trustscore.md) - [alibi.datasets](https://docs.seldon.ai/alibi-explain/api-reference/datasets.md) - [alibi.datasets.default](https://docs.seldon.ai/alibi-explain/api-reference/datasets/default.md) - [alibi.datasets.tensorflow](https://docs.seldon.ai/alibi-explain/api-reference/datasets/tensorflow.md) - [alibi.exceptions](https://docs.seldon.ai/alibi-explain/api-reference/exceptions.md) - [alibi.explainers](https://docs.seldon.ai/alibi-explain/api-reference/explainers.md) - [alibi.explainers.ale](https://docs.seldon.ai/alibi-explain/api-reference/explainers/ale.md) - [alibi.explainers.anchors](https://docs.seldon.ai/alibi-explain/api-reference/explainers/anchors.md) - [alibi.explainers.anchors.anchor\_base](https://docs.seldon.ai/alibi-explain/api-reference/explainers/anchors/anchor_base.md) - [alibi.explainers.anchors.anchor\_explanation](https://docs.seldon.ai/alibi-explain/api-reference/explainers/anchors/anchor_explanation.md) - [alibi.explainers.anchors.anchor\_image](https://docs.seldon.ai/alibi-explain/api-reference/explainers/anchors/anchor_image.md) - [alibi.explainers.anchors.anchor\_tabular](https://docs.seldon.ai/alibi-explain/api-reference/explainers/anchors/anchor_tabular.md) - [alibi.explainers.anchors.anchor\_tabular\_distributed](https://docs.seldon.ai/alibi-explain/api-reference/explainers/anchors/anchor_tabular_distributed.md) - [alibi.explainers.anchors.anchor\_text](https://docs.seldon.ai/alibi-explain/api-reference/explainers/anchors/anchor_text.md) - [alibi.explainers.anchors.language\_model\_text\_sampler](https://docs.seldon.ai/alibi-explain/api-reference/explainers/anchors/language_model_text_sampler.md) - [alibi.explainers.anchors.text\_samplers](https://docs.seldon.ai/alibi-explain/api-reference/explainers/anchors/text_samplers.md) - [alibi.explainers.backends](https://docs.seldon.ai/alibi-explain/api-reference/explainers/backends.md) - [alibi.explainers.backends.cfrl\_base](https://docs.seldon.ai/alibi-explain/api-reference/explainers/backends/cfrl_base.md) - [alibi.explainers.backends.cfrl\_tabular](https://docs.seldon.ai/alibi-explain/api-reference/explainers/backends/cfrl_tabular.md) - [alibi.explainers.backends.pytorch](https://docs.seldon.ai/alibi-explain/api-reference/explainers/backends/pytorch.md) - [alibi.explainers.backends.pytorch.cfrl\_base](https://docs.seldon.ai/alibi-explain/api-reference/explainers/backends/pytorch/cfrl_base.md) - [alibi.explainers.backends.pytorch.cfrl\_tabular](https://docs.seldon.ai/alibi-explain/api-reference/explainers/backends/pytorch/cfrl_tabular.md) - [alibi.explainers.backends.tensorflow](https://docs.seldon.ai/alibi-explain/api-reference/explainers/backends/tensorflow.md) - [alibi.explainers.backends.tensorflow.cfrl\_base](https://docs.seldon.ai/alibi-explain/api-reference/explainers/backends/tensorflow/cfrl_base.md) - [alibi.explainers.backends.tensorflow.cfrl\_tabular](https://docs.seldon.ai/alibi-explain/api-reference/explainers/backends/tensorflow/cfrl_tabular.md) - [alibi.explainers.cem](https://docs.seldon.ai/alibi-explain/api-reference/explainers/cem.md) - [alibi.explainers.cfproto](https://docs.seldon.ai/alibi-explain/api-reference/explainers/cfproto.md) - [alibi.explainers.cfrl\_base](https://docs.seldon.ai/alibi-explain/api-reference/explainers/cfrl_base.md) - [alibi.explainers.cfrl\_tabular](https://docs.seldon.ai/alibi-explain/api-reference/explainers/cfrl_tabular.md) - [alibi.explainers.counterfactual](https://docs.seldon.ai/alibi-explain/api-reference/explainers/counterfactual.md) - [alibi.explainers.integrated\_gradients](https://docs.seldon.ai/alibi-explain/api-reference/explainers/integrated_gradients.md) - [alibi.explainers.partial\_dependence](https://docs.seldon.ai/alibi-explain/api-reference/explainers/partial_dependence.md) - [alibi.explainers.pd\_variance](https://docs.seldon.ai/alibi-explain/api-reference/explainers/pd_variance.md) - [alibi.explainers.permutation\_importance](https://docs.seldon.ai/alibi-explain/api-reference/explainers/permutation_importance.md) - [alibi.explainers.shap\_wrappers](https://docs.seldon.ai/alibi-explain/api-reference/explainers/shap_wrappers.md) - [alibi.explainers.similarity](https://docs.seldon.ai/alibi-explain/api-reference/explainers/similarity.md) - [alibi.explainers.similarity.backends](https://docs.seldon.ai/alibi-explain/api-reference/explainers/similarity/backends.md) - [alibi.explainers.similarity.backends.pytorch](https://docs.seldon.ai/alibi-explain/api-reference/explainers/similarity/backends/pytorch.md) - [alibi.explainers.similarity.backends.pytorch.base](https://docs.seldon.ai/alibi-explain/api-reference/explainers/similarity/backends/pytorch/base.md) - [alibi.explainers.similarity.backends.tensorflow](https://docs.seldon.ai/alibi-explain/api-reference/explainers/similarity/backends/tensorflow.md) - [alibi.explainers.similarity.backends.tensorflow.base](https://docs.seldon.ai/alibi-explain/api-reference/explainers/similarity/backends/tensorflow/base.md) - [alibi.explainers.similarity.base](https://docs.seldon.ai/alibi-explain/api-reference/explainers/similarity/base.md) - [alibi.explainers.similarity.grad](https://docs.seldon.ai/alibi-explain/api-reference/explainers/similarity/grad.md) - [alibi.explainers.similarity.metrics](https://docs.seldon.ai/alibi-explain/api-reference/explainers/similarity/metrics.md) - [alibi.models](https://docs.seldon.ai/alibi-explain/api-reference/models.md) - [alibi.models.pytorch](https://docs.seldon.ai/alibi-explain/api-reference/models/pytorch.md) - [alibi.models.pytorch.actor\_critic](https://docs.seldon.ai/alibi-explain/api-reference/models/pytorch/actor_critic.md) - [alibi.models.pytorch.autoencoder](https://docs.seldon.ai/alibi-explain/api-reference/models/pytorch/autoencoder.md) - [alibi.models.pytorch.cfrl\_models](https://docs.seldon.ai/alibi-explain/api-reference/models/pytorch/cfrl_models.md) - [alibi.models.pytorch.metrics](https://docs.seldon.ai/alibi-explain/api-reference/models/pytorch/metrics.md) - [alibi.models.pytorch.model](https://docs.seldon.ai/alibi-explain/api-reference/models/pytorch/model.md) - [alibi.models.tensorflow](https://docs.seldon.ai/alibi-explain/api-reference/models/tensorflow.md) - [alibi.models.tensorflow.actor\_critic](https://docs.seldon.ai/alibi-explain/api-reference/models/tensorflow/actor_critic.md) - [alibi.models.tensorflow.autoencoder](https://docs.seldon.ai/alibi-explain/api-reference/models/tensorflow/autoencoder.md) - [alibi.models.tensorflow.cfrl\_models](https://docs.seldon.ai/alibi-explain/api-reference/models/tensorflow/cfrl_models.md) - [alibi.prototypes](https://docs.seldon.ai/alibi-explain/api-reference/prototypes.md) - [alibi.prototypes.protoselect](https://docs.seldon.ai/alibi-explain/api-reference/prototypes/protoselect.md) - [alibi.saving](https://docs.seldon.ai/alibi-explain/api-reference/saving.md) - [alibi.utils](https://docs.seldon.ai/alibi-explain/api-reference/utils.md) - [alibi.utils.approximation\_methods](https://docs.seldon.ai/alibi-explain/api-reference/utils/approximation_methods.md) - [alibi.utils.data](https://docs.seldon.ai/alibi-explain/api-reference/utils/data.md) - [alibi.utils.discretizer](https://docs.seldon.ai/alibi-explain/api-reference/utils/discretizer.md) - [alibi.utils.distance](https://docs.seldon.ai/alibi-explain/api-reference/utils/distance.md) - [alibi.utils.distributed](https://docs.seldon.ai/alibi-explain/api-reference/utils/distributed.md) - [alibi.utils.distributions](https://docs.seldon.ai/alibi-explain/api-reference/utils/distributions.md) - [alibi.utils.download](https://docs.seldon.ai/alibi-explain/api-reference/utils/download.md) - [alibi.utils.frameworks](https://docs.seldon.ai/alibi-explain/api-reference/utils/frameworks.md) - [alibi.utils.gradients](https://docs.seldon.ai/alibi-explain/api-reference/utils/gradients.md) - [alibi.utils.kernel](https://docs.seldon.ai/alibi-explain/api-reference/utils/kernel.md) - [alibi.utils.lang\_model](https://docs.seldon.ai/alibi-explain/api-reference/utils/lang_model.md) - [alibi.utils.mapping](https://docs.seldon.ai/alibi-explain/api-reference/utils/mapping.md) - [alibi.utils.missing\_optional\_dependency](https://docs.seldon.ai/alibi-explain/api-reference/utils/missing_optional_dependency.md) - [alibi.utils.tf](https://docs.seldon.ai/alibi-explain/api-reference/utils/tf.md) - [alibi.utils.visualization](https://docs.seldon.ai/alibi-explain/api-reference/utils/visualization.md) - [alibi.utils.wrappers](https://docs.seldon.ai/alibi-explain/api-reference/utils/wrappers.md) - [alibi.version](https://docs.seldon.ai/alibi-explain/api-reference/version.md) --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation This site has an agentic ask interface you may use to query the documentation dynamically by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://docs.seldon.ai/alibi-explain/index.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.