Search results

<code>KernelShap</code> returns no explanations when <code>link='logit'</code> and predicted proba is 0 or 1

Effort: S

Priority: Low

Type: EnhancementA better way of doing thingsA better way of doing things

Status: Open.

#952 In SeldonIO/alibi;

· ascillitoe opened on Jul 17, 2023

<code>GradSim</code> Improvements

Type: EnhancementA better way of doing thingsA better way of doing things

Status: Open.

#837 In SeldonIO/alibi;

· mauicv opened on Dec 13, 2022

Add reverse index for <code>extra_requires</code>

dependenciesPull requests that update a dependency filePull requests that update a dependency file

Priority: Low

Type: EnhancementA better way of doing thingsA better way of doing things

Status: Open.

#648 In SeldonIO/alibi;

· mauicv opened on Apr 29, 2022

Add Pandas Support For Explainers

Type: DesignDesign discussions and proposalsDesign discussions and proposals

Type: EnhancementA better way of doing thingsA better way of doing things

Status: Open.

#597 In SeldonIO/alibi;

· mauicv opened on Feb 16, 2022

Have a factory based creation of alibi explainers (based on config)

Type: EnhancementA better way of doing thingsA better way of doing things

Status: Open.

#494 In SeldonIO/alibi;

· sakoush opened on Oct 4, 2021

Allow <code>IntegratedGradients.target</code> to be inferred for classification models with class score

Priority: Medium

Type: APIAnything affecting the public API (what is exposed, function signatures, return values etc.)Anything affecting the public API (what is exposed, function signatures, return values etc.)

Type: EnhancementA better way of doing thingsA better way of doing things

Status: Open.

#492 In SeldonIO/alibi;

· sakoush opened on Oct 4, 2021

Validate explain parameters

Priority: Medium

Type: EnhancementA better way of doing thingsA better way of doing things

Status: Open.

#488 In SeldonIO/alibi;

· sakoush opened on Sep 24, 2021

Select subset of features for perturbations [Counterfactuals Guided by Prototypes]

BlockedIssue is blocked by some bigger issueIssue is blocked by some bigger issue

Type: BugSomething isn't workingSomething isn't working

Type: EnhancementA better way of doing thingsA better way of doing things

Type: Method extensionExtensions to existing methodsExtensions to existing methods

Status: Open.

#327 In SeldonIO/alibi;

· gibranhasan14 opened on Dec 8, 2020

IG error in TF 2.3.0 (Supporting TF models with different input types)

Type: EnhancementA better way of doing thingsA better way of doing things

Status: Open.

#285 In SeldonIO/alibi;

· Realvincentyuan opened on Aug 6, 2020

KernelSHAP raises a warning if output dim is 1 and task is regression

Type: EnhancementA better way of doing thingsA better way of doing things

Status: Open.

#226 In SeldonIO/alibi;

· jklaise opened on May 21, 2020

KernelSHAP argmax prediction for binary classifiers with one output

Type: BugSomething isn't workingSomething isn't working

Type: EnhancementA better way of doing thingsA better way of doing things

Status: Open.

#225 In SeldonIO/alibi;

· jklaise opened on May 21, 2020

Use numpy random generators to enable pickling of random state

Priority: Low

Type: EnhancementA better way of doing thingsA better way of doing things

Status: Open.

#209 In SeldonIO/alibi;

· jklaise opened on May 4, 2020