Model coverage

Family	Models	Differentiated output
Linear regression	LinearRegression, Ridge, Lasso, ElasticNet	Prediction
Linear classification	LogisticRegression, RidgeClassifier, LinearSVC	Decision score
Linear support-vector regression	LinearSVR	Prediction
Kernel support-vector regression	SVR, NuSVR	Prediction
Binary kernel classification	SVC, NuSVC	Decision score
Neural-network regression	MLPRegressor with squared-error or Poisson loss	Prediction, including the Poisson exponential output link
Neural-network classification	MLPClassifier	Binary or multiclass logits before logistic/softmax
Continuous pipelines	Supported scalers, polynomial expansion, PCA, and fitted feature selectors, then any supported estimator	Final estimator output, differentiated with respect to pipeline input features

MLP hidden activations may be identity, logistic, tanh, or ReLU. At ReLU’s nondifferentiable origin, skgrad uses a zero derivative, matching scikit-learn’s backpropagation convention. Scalar and multi-output regression, binary classification, and multiclass classification are supported.

Kernel SVMs support scikit-learn’s linear, poly, rbf, and sigmoid kernels. Multiclass kernel classifiers, callable kernels, and precomputed kernels are not currently supported.

Tree models are intentionally excluded. Their predictions are piecewise constant, so ordinary gradients are zero almost everywhere and undefined at split boundaries. Use TreeIG, which computes exact Integrated Gradients from the prediction jumps at tree split crossings.

skgrad expects finite dense numeric inputs. Sequential and nested sklearn pipelines may combine StandardScaler, RobustScaler, MaxAbsScaler, MinMaxScaler, PolynomialFeatures, PCA, and supported fitted feature selectors before a supported estimator. Gradients refer to the inputs of the supplied pipeline, including every supported preprocessing chain rule. Unknown transformers reject the entire analytic route; preprocessing is never silently removed. See pipeline conventions for clipping, whitening, feature selection, and attribution coordinates.

Important gaps

Coverage is focused rather than exhaustive. ColumnTransformer, categorical encoders, arbitrary FunctionTransformer functions, and unsupported custom transformer subclasses are not composed automatically. KernelRidge, Gaussian processes, generalized linear regressors such as PoissonRegressor, and many meta-estimators are also outside the current backend list. Having a coef_ attribute alone does not confer support.

Transformer recognition uses an explicit class list so an overridden transform method cannot silently inherit an incorrect derivative. Estimator subclasses are accepted only when they inherit the differentiated prediction method unchanged, including through intermediate base classes. Regressors protect predict; score classifiers protect decision_function. MLP classifiers protect predict and predict_proba because their exposed logits derive from that prediction path. Overrides of these methods are refused. This guard does not certify arbitrary changes to fitting or private helpers.

Nonlinear kernel SVMs depend on sklearn’s private fitted _gamma attribute to resolve gamma="scale" and gamma="auto". If it is unavailable, evaluation raises a clear compatibility error. CI tests scikit-learn pre-releases without imposing an upper dependency bound.

See pipeline gradients for the exact supported transformations and numerical fallback policy for unsupported models.