agentdb 1.6.1 → 2.0.0-alpha.2

package/dist/malp/models/malp.d.ts

@@ -0,0 +1,116 @@
+/**
+ * MALP Regressor - Maximum Agreement Linear Predictor
+ *
+ * A linear regression model that maximizes the Concordance Correlation Coefficient (CCC)
+ * instead of minimizing squared error. Provides better agreement between predictions
+ * and observations, especially for measurement validation and method comparison.
+ */
+import { Vector, Matrix } from '../core/linalg.js';
+import { OptimizerConfig, OptimizationResult } from '../core/optimizer.js';
+/**
+ * MALP model configuration
+ */
+export interface MALPConfig {
+    /** Whether to fit an intercept term */
+    fitIntercept: boolean;
+    /** Optimizer configuration */
+    optimizer: Partial<OptimizerConfig>;
+    /** Initialize with least squares solution */
+    useLeastSquaresInit: boolean;
+}
+/**
+ * Default MALP configuration
+ */
+export declare const DEFAULT_MALP_CONFIG: MALPConfig;
+/**
+ * MALP Regressor Model
+ *
+ * Fits a linear model by maximizing the Concordance Correlation Coefficient.
+ *
+ * Usage:
+ * ```typescript
+ * const model = new MALPRegressor();
+ * model.fit(X, y);
+ * const predictions = model.predict(X);
+ * const score = model.score(X, y);
+ * ```
+ */
+export declare class MALPRegressor {
+    private config;
+    private coefficients;
+    private intercept;
+    private isFitted;
+    private optimizationResult;
+    constructor(config?: Partial<MALPConfig>);
+    /**
+     * Fit the MALP model to training data
+     *
+     * @param X - Feature matrix (n_samples × n_features)
+     * @param y - Target values (n_samples)
+     * @returns this (for method chaining)
+     */
+    fit(X: number[][] | Matrix, y: number[] | Vector): this;
+    /**
+     * Make predictions using the fitted model
+     *
+     * @param X - Feature matrix (n_samples × n_features)
+     * @returns Predicted values (n_samples)
+     */
+    predict(X: number[][] | Matrix): number[];
+    /**
+     * Compute CCC score on test data
+     *
+     * @param X - Feature matrix (n_samples × n_features)
+     * @param y - True target values (n_samples)
+     * @returns CCC score (higher is better, range [-1, 1])
+     */
+    score(X: number[][] | Matrix, y: number[] | Vector): number;
+    /**
+     * Get model coefficients (excluding intercept)
+     */
+    getCoefficients(): number[] | null;
+    /**
+     * Get model intercept
+     */
+    getIntercept(): number;
+    /**
+     * Check if model is fitted
+     */
+    isTrained(): boolean;
+    /**
+     * Get optimization result details
+     */
+    getOptimizationResult(): OptimizationResult | null;
+    /**
+     * Get model summary statistics
+     */
+    summary(): {
+        coefficients: number[] | null;
+        intercept: number;
+        isFitted: boolean;
+        finalCCC: number | null;
+        iterations: number | null;
+        converged: boolean | null;
+    };
+    /**
+     * Reset the model to unfitted state
+     */
+    reset(): void;
+    /**
+     * Clone the model configuration (not the fitted state)
+     */
+    clone(): MALPRegressor;
+    /**
+     * Set model parameters directly (for advanced use cases)
+     */
+    setParams(coefficients: number[], intercept?: number): void;
+    /**
+     * Get feature importance based on coefficient magnitudes
+     */
+    getFeatureImportance(): number[] | null;
+}
+/**
+ * Convenience function for fitting MALP model
+ */
+export declare function fitMALP(X: number[][] | Matrix, y: number[] | Vector, config?: Partial<MALPConfig>): MALPRegressor;
+//# sourceMappingURL=malp.d.ts.map

package/dist/malp/models/malp.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"malp.d.ts","sourceRoot":"","sources":["../../../src/malp/models/malp.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG;AAEH,OAAO,EAAE,MAAM,EAAE,MAAM,EAA0E,MAAM,mBAAmB,CAAC;AAE3H,OAAO,EAA2B,eAAe,EAAE,kBAAkB,EAA4B,MAAM,sBAAsB,CAAC;AAE9H;;GAEG;AACH,MAAM,WAAW,UAAU;IACzB,uCAAuC;IACvC,YAAY,EAAE,OAAO,CAAC;IACtB,8BAA8B;IAC9B,SAAS,EAAE,OAAO,CAAC,eAAe,CAAC,CAAC;IACpC,6CAA6C;IAC7C,mBAAmB,EAAE,OAAO,CAAC;CAC9B;AAED;;GAEG;AACH,eAAO,MAAM,mBAAmB,EAAE,UAIjC,CAAC;AAEF;;;;;;;;;;;;GAYG;AACH,qBAAa,aAAa;IACxB,OAAO,CAAC,MAAM,CAAa;IAC3B,OAAO,CAAC,YAAY,CAAuB;IAC3C,OAAO,CAAC,SAAS,CAAa;IAC9B,OAAO,CAAC,QAAQ,CAAkB;IAClC,OAAO,CAAC,kBAAkB,CAAmC;gBAEjD,MAAM,GAAE,OAAO,CAAC,UAAU,CAAM;IAQ5C;;;;;;OAMG;IACH,GAAG,CAAC,CAAC,EAAE,MAAM,EAAE,EAAE,GAAG,MAAM,EAAE,CAAC,EAAE,MAAM,EAAE,GAAG,MAAM,GAAG,IAAI;IA2CvD;;;;;OAKG;IACH,OAAO,CAAC,CAAC,EAAE,MAAM,EAAE,EAAE,GAAG,MAAM,GAAG,MAAM,EAAE;IAuBzC;;;;;;OAMG;IACH,KAAK,CAAC,CAAC,EAAE,MAAM,EAAE,EAAE,GAAG,MAAM,EAAE,CAAC,EAAE,MAAM,EAAE,GAAG,MAAM,GAAG,MAAM;IAS3D;;OAEG;IACH,eAAe,IAAI,MAAM,EAAE,GAAG,IAAI;IAIlC;;OAEG;IACH,YAAY,IAAI,MAAM;IAItB;;OAEG;IACH,SAAS,IAAI,OAAO;IAIpB;;OAEG;IACH,qBAAqB,IAAI,kBAAkB,GAAG,IAAI;IAIlD;;OAEG;IACH,OAAO,IAAI;QACT,YAAY,EAAE,MAAM,EAAE,GAAG,IAAI,CAAC;QAC9B,SAAS,EAAE,MAAM,CAAC;QAClB,QAAQ,EAAE,OAAO,CAAC;QAClB,QAAQ,EAAE,MAAM,GAAG,IAAI,CAAC;QACxB,UAAU,EAAE,MAAM,GAAG,IAAI,CAAC;QAC1B,SAAS,EAAE,OAAO,GAAG,IAAI,CAAC;KAC3B;IAWD;;OAEG;IACH,KAAK,IAAI,IAAI;IAOb;;OAEG;IACH,KAAK,IAAI,aAAa;IAItB;;OAEG;IACH,SAAS,CAAC,YAAY,EAAE,MAAM,EAAE,EAAE,SAAS,GAAE,MAAU,GAAG,IAAI;IAM9D;;OAEG;IACH,oBAAoB,IAAI,MAAM,EAAE,GAAG,IAAI;CAaxC;AAED;;GAEG;AACH,wBAAgB,OAAO,CACrB,CAAC,EAAE,MAAM,EAAE,EAAE,GAAG,MAAM,EACtB,CAAC,EAAE,MAAM,EAAE,GAAG,MAAM,EACpB,MAAM,CAAC,EAAE,OAAO,CAAC,UAAU,CAAC,GAC3B,aAAa,CAIf"}

package/dist/malp/models/malp.js

@@ -0,0 +1,206 @@
+/**
+ * MALP Regressor - Maximum Agreement Linear Predictor
+ *
+ * A linear regression model that maximizes the Concordance Correlation Coefficient (CCC)
+ * instead of minimizing squared error. Provides better agreement between predictions
+ * and observations, especially for measurement validation and method comparison.
+ */
+import { createVector, createMatrix, addIntercept, matVecMultiply, leastSquares } from '../core/linalg.js';
+import { computeCCC } from '../core/ccc.js';
+import { GradientAscentOptimizer, DEFAULT_OPTIMIZER_CONFIG } from '../core/optimizer.js';
+/**
+ * Default MALP configuration
+ */
+export const DEFAULT_MALP_CONFIG = {
+    fitIntercept: true,
+    optimizer: DEFAULT_OPTIMIZER_CONFIG,
+    useLeastSquaresInit: true
+};
+/**
+ * MALP Regressor Model
+ *
+ * Fits a linear model by maximizing the Concordance Correlation Coefficient.
+ *
+ * Usage:
+ * ```typescript
+ * const model = new MALPRegressor();
+ * model.fit(X, y);
+ * const predictions = model.predict(X);
+ * const score = model.score(X, y);
+ * ```
+ */
+export class MALPRegressor {
+    config;
+    coefficients = null;
+    intercept = 0;
+    isFitted = false;
+    optimizationResult = null;
+    constructor(config = {}) {
+        this.config = {
+            ...DEFAULT_MALP_CONFIG,
+            ...config,
+            optimizer: { ...DEFAULT_OPTIMIZER_CONFIG, ...(config.optimizer || {}) }
+        };
+    }
+    /**
+     * Fit the MALP model to training data
+     *
+     * @param X - Feature matrix (n_samples × n_features)
+     * @param y - Target values (n_samples)
+     * @returns this (for method chaining)
+     */
+    fit(X, y) {
+        // Convert inputs to standard format
+        const XMatrix = Array.isArray(X) ? createMatrix(X) : X;
+        const yVector = Array.isArray(y) ? createVector(y) : y;
+        if (XMatrix.rows !== yVector.length) {
+            throw new Error(`Sample size mismatch: ${XMatrix.rows} !== ${yVector.length}`);
+        }
+        // Add intercept column if needed
+        const XWithIntercept = this.config.fitIntercept ? addIntercept(XMatrix) : XMatrix;
+        // Initialize parameters
+        let initialParams;
+        if (this.config.useLeastSquaresInit) {
+            // Initialize with least squares solution
+            initialParams = leastSquares(XWithIntercept, yVector);
+        }
+        else {
+            // Initialize with zeros
+            initialParams = createVector(new Array(XWithIntercept.cols).fill(0));
+        }
+        // Optimize parameters to maximize CCC
+        const optimizer = new GradientAscentOptimizer(this.config.optimizer);
+        this.optimizationResult = optimizer.optimize(XWithIntercept, yVector, initialParams);
+        // Extract coefficients and intercept
+        const params = this.optimizationResult.parameters;
+        if (this.config.fitIntercept) {
+            this.intercept = params.data[0];
+            this.coefficients = createVector(params.data.slice(1));
+        }
+        else {
+            this.intercept = 0;
+            this.coefficients = params;
+        }
+        this.isFitted = true;
+        return this;
+    }
+    /**
+     * Make predictions using the fitted model
+     *
+     * @param X - Feature matrix (n_samples × n_features)
+     * @returns Predicted values (n_samples)
+     */
+    predict(X) {
+        if (!this.isFitted || !this.coefficients) {
+            throw new Error('Model must be fitted before making predictions');
+        }
+        const XMatrix = Array.isArray(X) ? createMatrix(X) : X;
+        if (XMatrix.cols !== this.coefficients.length) {
+            throw new Error(`Feature dimension mismatch: ${XMatrix.cols} !== ${this.coefficients.length}`);
+        }
+        // Compute predictions: y_pred = X * β + intercept
+        const predictions = matVecMultiply(XMatrix, this.coefficients);
+        if (this.config.fitIntercept) {
+            for (let i = 0; i < predictions.length; i++) {
+                predictions.data[i] += this.intercept;
+            }
+        }
+        return predictions.data;
+    }
+    /**
+     * Compute CCC score on test data
+     *
+     * @param X - Feature matrix (n_samples × n_features)
+     * @param y - True target values (n_samples)
+     * @returns CCC score (higher is better, range [-1, 1])
+     */
+    score(X, y) {
+        const predictions = this.predict(X);
+        const yVector = Array.isArray(y) ? createVector(y) : y;
+        const predVector = createVector(predictions);
+        const { ccc } = computeCCC(yVector, predVector);
+        return ccc;
+    }
+    /**
+     * Get model coefficients (excluding intercept)
+     */
+    getCoefficients() {
+        return this.coefficients ? [...this.coefficients.data] : null;
+    }
+    /**
+     * Get model intercept
+     */
+    getIntercept() {
+        return this.intercept;
+    }
+    /**
+     * Check if model is fitted
+     */
+    isTrained() {
+        return this.isFitted;
+    }
+    /**
+     * Get optimization result details
+     */
+    getOptimizationResult() {
+        return this.optimizationResult ? { ...this.optimizationResult } : null;
+    }
+    /**
+     * Get model summary statistics
+     */
+    summary() {
+        return {
+            coefficients: this.getCoefficients(),
+            intercept: this.intercept,
+            isFitted: this.isFitted,
+            finalCCC: this.optimizationResult?.finalCCC ?? null,
+            iterations: this.optimizationResult?.iterations ?? null,
+            converged: this.optimizationResult?.converged ?? null
+        };
+    }
+    /**
+     * Reset the model to unfitted state
+     */
+    reset() {
+        this.coefficients = null;
+        this.intercept = 0;
+        this.isFitted = false;
+        this.optimizationResult = null;
+    }
+    /**
+     * Clone the model configuration (not the fitted state)
+     */
+    clone() {
+        return new MALPRegressor(this.config);
+    }
+    /**
+     * Set model parameters directly (for advanced use cases)
+     */
+    setParams(coefficients, intercept = 0) {
+        this.coefficients = createVector(coefficients);
+        this.intercept = intercept;
+        this.isFitted = true;
+    }
+    /**
+     * Get feature importance based on coefficient magnitudes
+     */
+    getFeatureImportance() {
+        if (!this.coefficients) {
+            return null;
+        }
+        // Compute absolute values for importance
+        const importance = this.coefficients.data.map(Math.abs);
+        // Normalize to sum to 1
+        const total = importance.reduce((sum, val) => sum + val, 0);
+        return total > 0 ? importance.map(val => val / total) : importance;
+    }
+}
+/**
+ * Convenience function for fitting MALP model
+ */
+export function fitMALP(X, y, config) {
+    const model = new MALPRegressor(config);
+    model.fit(X, y);
+    return model;
+}
+//# sourceMappingURL=malp.js.map

package/dist/malp/models/malp.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"malp.js","sourceRoot":"","sources":["../../../src/malp/models/malp.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG;AAEH,OAAO,EAAkB,YAAY,EAAE,YAAY,EAAE,YAAY,EAAE,cAAc,EAAE,YAAY,EAAE,MAAM,mBAAmB,CAAC;AAC3H,OAAO,EAAE,UAAU,EAAE,MAAM,gBAAgB,CAAC;AAC5C,OAAO,EAAE,uBAAuB,EAAuC,wBAAwB,EAAE,MAAM,sBAAsB,CAAC;AAc9H;;GAEG;AACH,MAAM,CAAC,MAAM,mBAAmB,GAAe;IAC7C,YAAY,EAAE,IAAI;IAClB,SAAS,EAAE,wBAAwB;IACnC,mBAAmB,EAAE,IAAI;CAC1B,CAAC;AAEF;;;;;;;;;;;;GAYG;AACH,MAAM,OAAO,aAAa;IAChB,MAAM,CAAa;IACnB,YAAY,GAAkB,IAAI,CAAC;IACnC,SAAS,GAAW,CAAC,CAAC;IACtB,QAAQ,GAAY,KAAK,CAAC;IAC1B,kBAAkB,GAA8B,IAAI,CAAC;IAE7D,YAAY,SAA8B,EAAE;QAC1C,IAAI,CAAC,MAAM,GAAG;YACZ,GAAG,mBAAmB;YACtB,GAAG,MAAM;YACT,SAAS,EAAE,EAAE,GAAG,wBAAwB,EAAE,GAAG,CAAC,MAAM,CAAC,SAAS,IAAI,EAAE,CAAC,EAAE;SACxE,CAAC;IACJ,CAAC;IAED;;;;;;OAMG;IACH,GAAG,CAAC,CAAsB,EAAE,CAAoB;QAC9C,oCAAoC;QACpC,MAAM,OAAO,GAAG,KAAK,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,YAAY,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;QACvD,MAAM,OAAO,GAAG,KAAK,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,YAAY,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;QAEvD,IAAI,OAAO,CAAC,IAAI,KAAK,OAAO,CAAC,MAAM,EAAE,CAAC;YACpC,MAAM,IAAI,KAAK,CAAC,yBAAyB,OAAO,CAAC,IAAI,QAAQ,OAAO,CAAC,MAAM,EAAE,CAAC,CAAC;QACjF,CAAC;QAED,iCAAiC;QACjC,MAAM,cAAc,GAAG,IAAI,CAAC,MAAM,CAAC,YAAY,CAAC,CAAC,CAAC,YAAY,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC;QAElF,wBAAwB;QACxB,IAAI,aAAqB,CAAC;QAE1B,IAAI,IAAI,CAAC,MAAM,CAAC,mBAAmB,EAAE,CAAC;YACpC,yCAAyC;YACzC,aAAa,GAAG,YAAY,CAAC,cAAc,EAAE,OAAO,CAAC,CAAC;QACxD,CAAC;aAAM,CAAC;YACN,wBAAwB;YACxB,aAAa,GAAG,YAAY,CAAC,IAAI,KAAK,CAAC,cAAc,CAAC,IAAI,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC,CAAC;QACvE,CAAC;QAED,sCAAsC;QACtC,MAAM,SAAS,GAAG,IAAI,uBAAuB,CAAC,IAAI,CAAC,MAAM,CAAC,SAAS,CAAC,CAAC;QACrE,IAAI,CAAC,kBAAkB,GAAG,SAAS,CAAC,QAAQ,CAAC,cAAc,EAAE,OAAO,EAAE,aAAa,CAAC,CAAC;QAErF,qCAAqC;QACrC,MAAM,MAAM,GAAG,IAAI,CAAC,kBAAkB,CAAC,UAAU,CAAC;QAElD,IAAI,IAAI,CAAC,MAAM,CAAC,YAAY,EAAE,CAAC;YAC7B,IAAI,CAAC,SAAS,GAAG,MAAM,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC;YAChC,IAAI,CAAC,YAAY,GAAG,YAAY,CAAC,MAAM,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,CAAC;QACzD,CAAC;aAAM,CAAC;YACN,IAAI,CAAC,SAAS,GAAG,CAAC,CAAC;YACnB,IAAI,CAAC,YAAY,GAAG,MAAM,CAAC;QAC7B,CAAC;QAED,IAAI,CAAC,QAAQ,GAAG,IAAI,CAAC;QAErB,OAAO,IAAI,CAAC;IACd,CAAC;IAED;;;;;OAKG;IACH,OAAO,CAAC,CAAsB;QAC5B,IAAI,CAAC,IAAI,CAAC,QAAQ,IAAI,CAAC,IAAI,CAAC,YAAY,EAAE,CAAC;YACzC,MAAM,IAAI,KAAK,CAAC,gDAAgD,CAAC,CAAC;QACpE,CAAC;QAED,MAAM,OAAO,GAAG,KAAK,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,YAAY,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;QAEvD,IAAI,OAAO,CAAC,IAAI,KAAK,IAAI,CAAC,YAAY,CAAC,MAAM,EAAE,CAAC;YAC9C,MAAM,IAAI,KAAK,CAAC,+BAA+B,OAAO,CAAC,IAAI,QAAQ,IAAI,CAAC,YAAY,CAAC,MAAM,EAAE,CAAC,CAAC;QACjG,CAAC;QAED,kDAAkD;QAClD,MAAM,WAAW,GAAG,cAAc,CAAC,OAAO,EAAE,IAAI,CAAC,YAAY,CAAC,CAAC;QAE/D,IAAI,IAAI,CAAC,MAAM,CAAC,YAAY,EAAE,CAAC;YAC7B,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,WAAW,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;gBAC5C,WAAW,CAAC,IAAI,CAAC,CAAC,CAAC,IAAI,IAAI,CAAC,SAAS,CAAC;YACxC,CAAC;QACH,CAAC;QAED,OAAO,WAAW,CAAC,IAAI,CAAC;IAC1B,CAAC;IAED;;;;;;OAMG;IACH,KAAK,CAAC,CAAsB,EAAE,CAAoB;QAChD,MAAM,WAAW,GAAG,IAAI,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC;QACpC,MAAM,OAAO,GAAG,KAAK,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,YAAY,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;QACvD,MAAM,UAAU,GAAG,YAAY,CAAC,WAAW,CAAC,CAAC;QAE7C,MAAM,EAAE,GAAG,EAAE,GAAG,UAAU,CAAC,OAAO,EAAE,UAAU,CAAC,CAAC;QAChD,OAAO,GAAG,CAAC;IACb,CAAC;IAED;;OAEG;IACH,eAAe;QACb,OAAO,IAAI,CAAC,YAAY,CAAC,CAAC,CAAC,CAAC,GAAG,IAAI,CAAC,YAAY,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC;IAChE,CAAC;IAED;;OAEG;IACH,YAAY;QACV,OAAO,IAAI,CAAC,SAAS,CAAC;IACxB,CAAC;IAED;;OAEG;IACH,SAAS;QACP,OAAO,IAAI,CAAC,QAAQ,CAAC;IACvB,CAAC;IAED;;OAEG;IACH,qBAAqB;QACnB,OAAO,IAAI,CAAC,kBAAkB,CAAC,CAAC,CAAC,EAAE,GAAG,IAAI,CAAC,kBAAkB,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC;IACzE,CAAC;IAED;;OAEG;IACH,OAAO;QAQL,OAAO;YACL,YAAY,EAAE,IAAI,CAAC,eAAe,EAAE;YACpC,SAAS,EAAE,IAAI,CAAC,SAAS;YACzB,QAAQ,EAAE,IAAI,CAAC,QAAQ;YACvB,QAAQ,EAAE,IAAI,CAAC,kBAAkB,EAAE,QAAQ,IAAI,IAAI;YACnD,UAAU,EAAE,IAAI,CAAC,kBAAkB,EAAE,UAAU,IAAI,IAAI;YACvD,SAAS,EAAE,IAAI,CAAC,kBAAkB,EAAE,SAAS,IAAI,IAAI;SACtD,CAAC;IACJ,CAAC;IAED;;OAEG;IACH,KAAK;QACH,IAAI,CAAC,YAAY,GAAG,IAAI,CAAC;QACzB,IAAI,CAAC,SAAS,GAAG,CAAC,CAAC;QACnB,IAAI,CAAC,QAAQ,GAAG,KAAK,CAAC;QACtB,IAAI,CAAC,kBAAkB,GAAG,IAAI,CAAC;IACjC,CAAC;IAED;;OAEG;IACH,KAAK;QACH,OAAO,IAAI,aAAa,CAAC,IAAI,CAAC,MAAM,CAAC,CAAC;IACxC,CAAC;IAED;;OAEG;IACH,SAAS,CAAC,YAAsB,EAAE,YAAoB,CAAC;QACrD,IAAI,CAAC,YAAY,GAAG,YAAY,CAAC,YAAY,CAAC,CAAC;QAC/C,IAAI,CAAC,SAAS,GAAG,SAAS,CAAC;QAC3B,IAAI,CAAC,QAAQ,GAAG,IAAI,CAAC;IACvB,CAAC;IAED;;OAEG;IACH,oBAAoB;QAClB,IAAI,CAAC,IAAI,CAAC,YAAY,EAAE,CAAC;YACvB,OAAO,IAAI,CAAC;QACd,CAAC;QAED,yCAAyC;QACzC,MAAM,UAAU,GAAG,IAAI,CAAC,YAAY,CAAC,IAAI,CAAC,GAAG,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC;QAExD,wBAAwB;QACxB,MAAM,KAAK,GAAG,UAAU,CAAC,MAAM,CAAC,CAAC,GAAG,EAAE,GAAG,EAAE,EAAE,CAAC,GAAG,GAAG,GAAG,EAAE,CAAC,CAAC,CAAC;QAE5D,OAAO,KAAK,GAAG,CAAC,CAAC,CAAC,CAAC,UAAU,CAAC,GAAG,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,UAAU,CAAC;IACrE,CAAC;CACF;AAED;;GAEG;AACH,MAAM,UAAU,OAAO,CACrB,CAAsB,EACtB,CAAoB,EACpB,MAA4B;IAE5B,MAAM,KAAK,GAAG,IAAI,aAAa,CAAC,MAAM,CAAC,CAAC;IACxC,KAAK,CAAC,GAAG,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC;IAChB,OAAO,KAAK,CAAC;AACf,CAAC"}

package/dist/malp/models/regressor.d.ts

@@ -0,0 +1,80 @@
+/**
+ * MALP Regressor - Maximum Agreement Linear Program
+ *
+ * Linear regression that maximizes concordance correlation coefficient
+ * instead of minimizing squared error.
+ */
+import { OptimizerConfig } from '../core/optimizer';
+export interface MALPRegressorConfig extends Partial<OptimizerConfig> {
+    fitIntercept?: boolean;
+    normalize?: boolean;
+    regularization?: number;
+}
+export type MALPConfig = MALPRegressorConfig;
+export declare const DEFAULT_MALP_CONFIG: Required<MALPRegressorConfig>;
+export interface FitResult {
+    weights: number[];
+    intercept: number;
+    ccc: number;
+    r2: number;
+    rmse: number;
+    mae: number;
+    iterations: number;
+    converged: boolean;
+}
+export declare class MALPRegressor {
+    private weights;
+    private intercept;
+    private optimizer;
+    private config;
+    private xMean;
+    private xStd;
+    private fitted;
+    constructor(config?: MALPRegressorConfig);
+    /**
+     * Fit MALP regressor to training data
+     * @param X Feature matrix (n_samples × n_features)
+     * @param y Target values (n_samples)
+     */
+    fit(X: number[][], y: number[]): FitResult;
+    /**
+     * Predict target values for new data
+     * @param X Feature matrix (n_samples × n_features)
+     */
+    predict(X: number[][]): number[];
+    /**
+     * Score the model on test data
+     * @param X Feature matrix
+     * @param y True target values
+     * @returns CCC score
+     */
+    score(X: number[][], y: number[]): number;
+    /**
+     * Get model coefficients
+     */
+    getCoefficients(): {
+        weights: number[];
+        intercept: number;
+    };
+    /**
+     * Normalize features (zero mean, unit variance)
+     */
+    private normalizeFeatures;
+    /**
+     * Compute evaluation metrics
+     */
+    private computeMetrics;
+}
+/**
+ * Cross-validation for MALP regressor
+ */
+export declare function crossValidateMALP(X: number[][], y: number[], nFolds?: number, config?: MALPRegressorConfig): {
+    meanCCC: number;
+    stdCCC: number;
+    foldScores: number[];
+};
+/**
+ * Convenience function to fit a MALP model
+ */
+export declare function fitMALP(X: number[][], y: number[], config?: MALPRegressorConfig): FitResult;
+//# sourceMappingURL=regressor.d.ts.map

package/dist/malp/models/regressor.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"regressor.d.ts","sourceRoot":"","sources":["../../../src/malp/models/regressor.ts"],"names":[],"mappings":"AAAA;;;;;GAKG;AAGH,OAAO,EAAiB,eAAe,EAAmB,MAAM,mBAAmB,CAAC;AAEpF,MAAM,WAAW,mBAAoB,SAAQ,OAAO,CAAC,eAAe,CAAC;IACnE,YAAY,CAAC,EAAE,OAAO,CAAC;IACvB,SAAS,CAAC,EAAE,OAAO,CAAC;IACpB,cAAc,CAAC,EAAE,MAAM,CAAC;CACzB;AAGD,MAAM,MAAM,UAAU,GAAG,mBAAmB,CAAC;AAE7C,eAAO,MAAM,mBAAmB,EAAE,QAAQ,CAAC,mBAAmB,CAU7D,CAAC;AAEF,MAAM,WAAW,SAAS;IACxB,OAAO,EAAE,MAAM,EAAE,CAAC;IAClB,SAAS,EAAE,MAAM,CAAC;IAClB,GAAG,EAAE,MAAM,CAAC;IACZ,EAAE,EAAE,MAAM,CAAC;IACX,IAAI,EAAE,MAAM,CAAC;IACb,GAAG,EAAE,MAAM,CAAC;IACZ,UAAU,EAAE,MAAM,CAAC;IACnB,SAAS,EAAE,OAAO,CAAC;CACpB;AAED,qBAAa,aAAa;IACxB,OAAO,CAAC,OAAO,CAAgB;IAC/B,OAAO,CAAC,SAAS,CAAa;IAC9B,OAAO,CAAC,SAAS,CAAgB;IACjC,OAAO,CAAC,MAAM,CAAgC;IAC9C,OAAO,CAAC,KAAK,CAAgB;IAC7B,OAAO,CAAC,IAAI,CAAgB;IAC5B,OAAO,CAAC,MAAM,CAAkB;gBAEpB,MAAM,GAAE,mBAAwB;IAuB5C;;;;OAIG;IACH,GAAG,CAAC,CAAC,EAAE,MAAM,EAAE,EAAE,EAAE,CAAC,EAAE,MAAM,EAAE,GAAG,SAAS;IAqD1C;;;OAGG;IACH,OAAO,CAAC,CAAC,EAAE,MAAM,EAAE,EAAE,GAAG,MAAM,EAAE;IA+BhC;;;;;OAKG;IACH,KAAK,CAAC,CAAC,EAAE,MAAM,EAAE,EAAE,EAAE,CAAC,EAAE,MAAM,EAAE,GAAG,MAAM;IAMzC;;OAEG;IACH,eAAe,IAAI;QAAE,OAAO,EAAE,MAAM,EAAE,CAAC;QAAC,SAAS,EAAE,MAAM,CAAA;KAAE;IAU3D;;OAEG;IACH,OAAO,CAAC,iBAAiB;IAsCzB;;OAEG;IACH,OAAO,CAAC,cAAc;CAyBvB;AAED;;GAEG;AACH,wBAAgB,iBAAiB,CAC/B,CAAC,EAAE,MAAM,EAAE,EAAE,EACb,CAAC,EAAE,MAAM,EAAE,EACX,MAAM,GAAE,MAAU,EAClB,MAAM,GAAE,mBAAwB,GAC/B;IACD,OAAO,EAAE,MAAM,CAAC;IAChB,MAAM,EAAE,MAAM,CAAC;IACf,UAAU,EAAE,MAAM,EAAE,CAAC;CACtB,CA2BA;AAED;;GAEG;AACH,wBAAgB,OAAO,CACrB,CAAC,EAAE,MAAM,EAAE,EAAE,EACb,CAAC,EAAE,MAAM,EAAE,EACX,MAAM,GAAE,mBAAwB,GAC/B,SAAS,CAGX"}

package/dist/malp/models/regressor.js

@@ -0,0 +1,229 @@
+/**
+ * MALP Regressor - Maximum Agreement Linear Program
+ *
+ * Linear regression that maximizes concordance correlation coefficient
+ * instead of minimizing squared error.
+ */
+import { calculateCCC } from '../core/ccc';
+import { MALPOptimizer } from '../core/optimizer';
+export const DEFAULT_MALP_CONFIG = {
+    learningRate: 0.01,
+    maxIterations: 1000,
+    tolerance: 1e-6,
+    momentum: 0.9,
+    adaptive: true,
+    verbose: false,
+    fitIntercept: true,
+    normalize: false,
+    regularization: 0.0
+};
+export class MALPRegressor {
+    weights = [];
+    intercept = 0;
+    optimizer;
+    config;
+    xMean = [];
+    xStd = [];
+    fitted = false;
+    constructor(config = {}) {
+        this.config = {
+            learningRate: config.learningRate ?? 0.01,
+            maxIterations: config.maxIterations ?? 1000,
+            tolerance: config.tolerance ?? 1e-6,
+            momentum: config.momentum ?? 0.9,
+            adaptive: config.adaptive ?? true,
+            verbose: config.verbose ?? false,
+            fitIntercept: config.fitIntercept ?? true,
+            normalize: config.normalize ?? false,
+            regularization: config.regularization ?? 0.0
+        };
+        this.optimizer = new MALPOptimizer({
+            learningRate: this.config.learningRate,
+            maxIterations: this.config.maxIterations,
+            tolerance: this.config.tolerance,
+            momentum: this.config.momentum,
+            adaptive: this.config.adaptive,
+            verbose: this.config.verbose
+        });
+    }
+    /**
+     * Fit MALP regressor to training data
+     * @param X Feature matrix (n_samples × n_features)
+     * @param y Target values (n_samples)
+     */
+    fit(X, y) {
+        if (X.length !== y.length) {
+            throw new Error(`Sample size mismatch: X has ${X.length}, y has ${y.length}`);
+        }
+        if (X.length === 0) {
+            throw new Error('Cannot fit with empty dataset');
+        }
+        // Normalize features if requested
+        let XProcessed = X;
+        if (this.config.normalize) {
+            const result = this.normalizeFeatures(X);
+            XProcessed = result.normalized;
+            this.xMean = result.mean;
+            this.xStd = result.std;
+        }
+        // Add intercept column if requested
+        if (this.config.fitIntercept) {
+            XProcessed = XProcessed.map(row => [1, ...row]);
+        }
+        // Optimize weights to maximize CCC
+        const optimResult = this.optimizer.optimize(XProcessed, y);
+        // Extract weights and intercept
+        if (this.config.fitIntercept) {
+            this.intercept = optimResult.weights[0];
+            this.weights = optimResult.weights.slice(1);
+        }
+        else {
+            this.intercept = 0;
+            this.weights = optimResult.weights;
+        }
+        this.fitted = true;
+        // Compute final metrics
+        const yPred = this.predict(X);
+        const metrics = this.computeMetrics(y, yPred);
+        return {
+            weights: this.weights,
+            intercept: this.intercept,
+            ccc: optimResult.finalCCC,
+            r2: metrics.r2,
+            rmse: metrics.rmse,
+            mae: metrics.mae,
+            iterations: optimResult.iterations,
+            converged: optimResult.converged
+        };
+    }
+    /**
+     * Predict target values for new data
+     * @param X Feature matrix (n_samples × n_features)
+     */
+    predict(X) {
+        if (!this.fitted) {
+            throw new Error('Model must be fitted before prediction');
+        }
+        if (X[0].length !== this.weights.length) {
+            throw new Error(`Feature dimension mismatch: expected ${this.weights.length}, got ${X[0].length}`);
+        }
+        return X.map(row => {
+            let xProcessed = row;
+            // Apply normalization if used during training
+            if (this.config.normalize) {
+                xProcessed = row.map((val, j) => (val - this.xMean[j]) / this.xStd[j]);
+            }
+            // Compute prediction
+            const prediction = xProcessed.reduce((sum, val, j) => sum + val * this.weights[j], this.intercept);
+            return prediction;
+        });
+    }
+    /**
+     * Score the model on test data
+     * @param X Feature matrix
+     * @param y True target values
+     * @returns CCC score
+     */
+    score(X, y) {
+        const yPred = this.predict(X);
+        const result = calculateCCC(yPred, y);
+        return result.ccc;
+    }
+    /**
+     * Get model coefficients
+     */
+    getCoefficients() {
+        if (!this.fitted) {
+            throw new Error('Model not fitted');
+        }
+        return {
+            weights: [...this.weights],
+            intercept: this.intercept
+        };
+    }
+    /**
+     * Normalize features (zero mean, unit variance)
+     */
+    normalizeFeatures(X) {
+        const nFeatures = X[0].length;
+        const mean = Array(nFeatures).fill(0);
+        const std = Array(nFeatures).fill(0);
+        // Compute means
+        for (const row of X) {
+            for (let j = 0; j < nFeatures; j++) {
+                mean[j] += row[j];
+            }
+        }
+        for (let j = 0; j < nFeatures; j++) {
+            mean[j] /= X.length;
+        }
+        // Compute standard deviations
+        for (const row of X) {
+            for (let j = 0; j < nFeatures; j++) {
+                std[j] += (row[j] - mean[j]) ** 2;
+            }
+        }
+        for (let j = 0; j < nFeatures; j++) {
+            std[j] = Math.sqrt(std[j] / X.length);
+            if (std[j] === 0)
+                std[j] = 1; // Prevent division by zero
+        }
+        // Normalize
+        const normalized = X.map(row => row.map((val, j) => (val - mean[j]) / std[j]));
+        return { normalized, mean, std };
+    }
+    /**
+     * Compute evaluation metrics
+     */
+    computeMetrics(yTrue, yPred) {
+        const n = yTrue.length;
+        const yMean = yTrue.reduce((sum, val) => sum + val, 0) / n;
+        let ssRes = 0;
+        let ssTot = 0;
+        let sumAbsError = 0;
+        for (let i = 0; i < n; i++) {
+            const error = yTrue[i] - yPred[i];
+            ssRes += error * error;
+            ssTot += (yTrue[i] - yMean) ** 2;
+            sumAbsError += Math.abs(error);
+        }
+        const r2 = 1 - (ssRes / ssTot);
+        const rmse = Math.sqrt(ssRes / n);
+        const mae = sumAbsError / n;
+        return { r2, rmse, mae };
+    }
+}
+/**
+ * Cross-validation for MALP regressor
+ */
+export function crossValidateMALP(X, y, nFolds = 5, config = {}) {
+    const n = X.length;
+    const foldSize = Math.floor(n / nFolds);
+    const scores = [];
+    for (let fold = 0; fold < nFolds; fold++) {
+        const testStart = fold * foldSize;
+        const testEnd = fold === nFolds - 1 ? n : testStart + foldSize;
+        // Split data
+        const XTrain = [...X.slice(0, testStart), ...X.slice(testEnd)];
+        const yTrain = [...y.slice(0, testStart), ...y.slice(testEnd)];
+        const XTest = X.slice(testStart, testEnd);
+        const yTest = y.slice(testStart, testEnd);
+        // Fit and score
+        const model = new MALPRegressor(config);
+        model.fit(XTrain, yTrain);
+        const score = model.score(XTest, yTest);
+        scores.push(score);
+    }
+    const meanCCC = scores.reduce((sum, s) => sum + s, 0) / nFolds;
+    const variance = scores.reduce((sum, s) => sum + (s - meanCCC) ** 2, 0) / nFolds;
+    const stdCCC = Math.sqrt(variance);
+    return { meanCCC, stdCCC, foldScores: scores };
+}
+/**
+ * Convenience function to fit a MALP model
+ */
+export function fitMALP(X, y, config = {}) {
+    const model = new MALPRegressor(config);
+    return model.fit(X, y);
+}
+//# sourceMappingURL=regressor.js.map

package/dist/malp/models/regressor.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"regressor.js","sourceRoot":"","sources":["../../../src/malp/models/regressor.ts"],"names":[],"mappings":"AAAA;;;;;GAKG;AAEH,OAAO,EAAE,YAAY,EAAa,MAAM,aAAa,CAAC;AACtD,OAAO,EAAE,aAAa,EAAoC,MAAM,mBAAmB,CAAC;AAWpF,MAAM,CAAC,MAAM,mBAAmB,GAAkC;IAChE,YAAY,EAAE,IAAI;IAClB,aAAa,EAAE,IAAI;IACnB,SAAS,EAAE,IAAI;IACf,QAAQ,EAAE,GAAG;IACb,QAAQ,EAAE,IAAI;IACd,OAAO,EAAE,KAAK;IACd,YAAY,EAAE,IAAI;IAClB,SAAS,EAAE,KAAK;IAChB,cAAc,EAAE,GAAG;CACpB,CAAC;AAaF,MAAM,OAAO,aAAa;IAChB,OAAO,GAAa,EAAE,CAAC;IACvB,SAAS,GAAW,CAAC,CAAC;IACtB,SAAS,CAAgB;IACzB,MAAM,CAAgC;IACtC,KAAK,GAAa,EAAE,CAAC;IACrB,IAAI,GAAa,EAAE,CAAC;IACpB,MAAM,GAAY,KAAK,CAAC;IAEhC,YAAY,SAA8B,EAAE;QAC1C,IAAI,CAAC,MAAM,GAAG;YACZ,YAAY,EAAE,MAAM,CAAC,YAAY,IAAI,IAAI;YACzC,aAAa,EAAE,MAAM,CAAC,aAAa,IAAI,IAAI;YAC3C,SAAS,EAAE,MAAM,CAAC,SAAS,IAAI,IAAI;YACnC,QAAQ,EAAE,MAAM,CAAC,QAAQ,IAAI,GAAG;YAChC,QAAQ,EAAE,MAAM,CAAC,QAAQ,IAAI,IAAI;YACjC,OAAO,EAAE,MAAM,CAAC,OAAO,IAAI,KAAK;YAChC,YAAY,EAAE,MAAM,CAAC,YAAY,IAAI,IAAI;YACzC,SAAS,EAAE,MAAM,CAAC,SAAS,IAAI,KAAK;YACpC,cAAc,EAAE,MAAM,CAAC,cAAc,IAAI,GAAG;SAC7C,CAAC;QAEF,IAAI,CAAC,SAAS,GAAG,IAAI,aAAa,CAAC;YACjC,YAAY,EAAE,IAAI,CAAC,MAAM,CAAC,YAAY;YACtC,aAAa,EAAE,IAAI,CAAC,MAAM,CAAC,aAAa;YACxC,SAAS,EAAE,IAAI,CAAC,MAAM,CAAC,SAAS;YAChC,QAAQ,EAAE,IAAI,CAAC,MAAM,CAAC,QAAQ;YAC9B,QAAQ,EAAE,IAAI,CAAC,MAAM,CAAC,QAAQ;YAC9B,OAAO,EAAE,IAAI,CAAC,MAAM,CAAC,OAAO;SAC7B,CAAC,CAAC;IACL,CAAC;IAED;;;;OAIG;IACH,GAAG,CAAC,CAAa,EAAE,CAAW;QAC5B,IAAI,CAAC,CAAC,MAAM,KAAK,CAAC,CAAC,MAAM,EAAE,CAAC;YAC1B,MAAM,IAAI,KAAK,CAAC,+BAA+B,CAAC,CAAC,MAAM,WAAW,CAAC,CAAC,MAAM,EAAE,CAAC,CAAC;QAChF,CAAC;QAED,IAAI,CAAC,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;YACnB,MAAM,IAAI,KAAK,CAAC,+BAA+B,CAAC,CAAC;QACnD,CAAC;QAED,kCAAkC;QAClC,IAAI,UAAU,GAAG,CAAC,CAAC;QACnB,IAAI,IAAI,CAAC,MAAM,CAAC,SAAS,EAAE,CAAC;YAC1B,MAAM,MAAM,GAAG,IAAI,CAAC,iBAAiB,CAAC,CAAC,CAAC,CAAC;YACzC,UAAU,GAAG,MAAM,CAAC,UAAU,CAAC;YAC/B,IAAI,CAAC,KAAK,GAAG,MAAM,CAAC,IAAI,CAAC;YACzB,IAAI,CAAC,IAAI,GAAG,MAAM,CAAC,GAAG,CAAC;QACzB,CAAC;QAED,oCAAoC;QACpC,IAAI,IAAI,CAAC,MAAM,CAAC,YAAY,EAAE,CAAC;YAC7B,UAAU,GAAG,UAAU,CAAC,GAAG,CAAC,GAAG,CAAC,EAAE,CAAC,CAAC,CAAC,EAAE,GAAG,GAAG,CAAC,CAAC,CAAC;QAClD,CAAC;QAED,mCAAmC;QACnC,MAAM,WAAW,GAAG,IAAI,CAAC,SAAS,CAAC,QAAQ,CAAC,UAAU,EAAE,CAAC,CAAC,CAAC;QAE3D,gCAAgC;QAChC,IAAI,IAAI,CAAC,MAAM,CAAC,YAAY,EAAE,CAAC;YAC7B,IAAI,CAAC,SAAS,GAAG,WAAW,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC;YACxC,IAAI,CAAC,OAAO,GAAG,WAAW,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC;QAC9C,CAAC;aAAM,CAAC;YACN,IAAI,CAAC,SAAS,GAAG,CAAC,CAAC;YACnB,IAAI,CAAC,OAAO,GAAG,WAAW,CAAC,OAAO,CAAC;QACrC,CAAC;QAED,IAAI,CAAC,MAAM,GAAG,IAAI,CAAC;QAEnB,wBAAwB;QACxB,MAAM,KAAK,GAAG,IAAI,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC;QAC9B,MAAM,OAAO,GAAG,IAAI,CAAC,cAAc,CAAC,CAAC,EAAE,KAAK,CAAC,CAAC;QAE9C,OAAO;YACL,OAAO,EAAE,IAAI,CAAC,OAAO;YACrB,SAAS,EAAE,IAAI,CAAC,SAAS;YACzB,GAAG,EAAE,WAAW,CAAC,QAAQ;YACzB,EAAE,EAAE,OAAO,CAAC,EAAE;YACd,IAAI,EAAE,OAAO,CAAC,IAAI;YAClB,GAAG,EAAE,OAAO,CAAC,GAAG;YAChB,UAAU,EAAE,WAAW,CAAC,UAAU;YAClC,SAAS,EAAE,WAAW,CAAC,SAAS;SACjC,CAAC;IACJ,CAAC;IAED;;;OAGG;IACH,OAAO,CAAC,CAAa;QACnB,IAAI,CAAC,IAAI,CAAC,MAAM,EAAE,CAAC;YACjB,MAAM,IAAI,KAAK,CAAC,wCAAwC,CAAC,CAAC;QAC5D,CAAC;QAED,IAAI,CAAC,CAAC,CAAC,CAAC,CAAC,MAAM,KAAK,IAAI,CAAC,OAAO,CAAC,MAAM,EAAE,CAAC;YACxC,MAAM,IAAI,KAAK,CACb,wCAAwC,IAAI,CAAC,OAAO,CAAC,MAAM,SAAS,CAAC,CAAC,CAAC,CAAC,CAAC,MAAM,EAAE,CAClF,CAAC;QACJ,CAAC;QAED,OAAO,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,EAAE;YACjB,IAAI,UAAU,GAAG,GAAG,CAAC;YAErB,8CAA8C;YAC9C,IAAI,IAAI,CAAC,MAAM,CAAC,SAAS,EAAE,CAAC;gBAC1B,UAAU,GAAG,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,EAAE,CAAC,EAAE,EAAE,CAC9B,CAAC,GAAG,GAAG,IAAI,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,GAAG,IAAI,CAAC,IAAI,CAAC,CAAC,CAAC,CACrC,CAAC;YACJ,CAAC;YAED,qBAAqB;YACrB,MAAM,UAAU,GAAG,UAAU,CAAC,MAAM,CAClC,CAAC,GAAG,EAAE,GAAG,EAAE,CAAC,EAAE,EAAE,CAAC,GAAG,GAAG,GAAG,GAAG,IAAI,CAAC,OAAO,CAAC,CAAC,CAAC,EAC5C,IAAI,CAAC,SAAS,CACf,CAAC;YAEF,OAAO,UAAU,CAAC;QACpB,CAAC,CAAC,CAAC;IACL,CAAC;IAED;;;;;OAKG;IACH,KAAK,CAAC,CAAa,EAAE,CAAW;QAC9B,MAAM,KAAK,GAAG,IAAI,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC;QAC9B,MAAM,MAAM,GAAG,YAAY,CAAC,KAAK,EAAE,CAAC,CAAC,CAAC;QACtC,OAAO,MAAM,CAAC,GAAG,CAAC;IACpB,CAAC;IAED;;OAEG;IACH,eAAe;QACb,IAAI,CAAC,IAAI,CAAC,MAAM,EAAE,CAAC;YACjB,MAAM,IAAI,KAAK,CAAC,kBAAkB,CAAC,CAAC;QACtC,CAAC;QACD,OAAO;YACL,OAAO,EAAE,CAAC,GAAG,IAAI,CAAC,OAAO,CAAC;YAC1B,SAAS,EAAE,IAAI,CAAC,SAAS;SAC1B,CAAC;IACJ,CAAC;IAED;;OAEG;IACK,iBAAiB,CAAC,CAAa;QAKrC,MAAM,SAAS,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC,MAAM,CAAC;QAC9B,MAAM,IAAI,GAAG,KAAK,CAAC,SAAS,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC;QACtC,MAAM,GAAG,GAAG,KAAK,CAAC,SAAS,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC;QAErC,gBAAgB;QAChB,KAAK,MAAM,GAAG,IAAI,CAAC,EAAE,CAAC;YACpB,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,SAAS,EAAE,CAAC,EAAE,EAAE,CAAC;gBACnC,IAAI,CAAC,CAAC,CAAC,IAAI,GAAG,CAAC,CAAC,CAAC,CAAC;YACpB,CAAC;QACH,CAAC;QACD,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,SAAS,EAAE,CAAC,EAAE,EAAE,CAAC;YACnC,IAAI,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,MAAM,CAAC;QACtB,CAAC;QAED,8BAA8B;QAC9B,KAAK,MAAM,GAAG,IAAI,CAAC,EAAE,CAAC;YACpB,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,SAAS,EAAE,CAAC,EAAE,EAAE,CAAC;gBACnC,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC,GAAG,IAAI,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC;YACpC,CAAC;QACH,CAAC;QACD,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,SAAS,EAAE,CAAC,EAAE,EAAE,CAAC;YACnC,GAAG,CAAC,CAAC,CAAC,GAAG,IAAI,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,MAAM,CAAC,CAAC;YACtC,IAAI,GAAG,CAAC,CAAC,CAAC,KAAK,CAAC;gBAAE,GAAG,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,2BAA2B;QAC3D,CAAC;QAED,YAAY;QACZ,MAAM,UAAU,GAAG,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,EAAE,CAC7B,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,EAAE,CAAC,EAAE,EAAE,CAAC,CAAC,GAAG,GAAG,IAAI,CAAC,CAAC,CAAC,CAAC,GAAG,GAAG,CAAC,CAAC,CAAC,CAAC,CAC9C,CAAC;QAEF,OAAO,EAAE,UAAU,EAAE,IAAI,EAAE,GAAG,EAAE,CAAC;IACnC,CAAC;IAED;;OAEG;IACK,cAAc,CAAC,KAAe,EAAE,KAAe;QAKrD,MAAM,CAAC,GAAG,KAAK,CAAC,MAAM,CAAC;QACvB,MAAM,KAAK,GAAG,KAAK,CAAC,MAAM,CAAC,CAAC,GAAG,EAAE,GAAG,EAAE,EAAE,CAAC,GAAG,GAAG,GAAG,EAAE,CAAC,CAAC,GAAG,CAAC,CAAC;QAE3D,IAAI,KAAK,GAAG,CAAC,CAAC;QACd,IAAI,KAAK,GAAG,CAAC,CAAC;QACd,IAAI,WAAW,GAAG,CAAC,CAAC;QAEpB,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC;YAC3B,MAAM,KAAK,GAAG,KAAK,CAAC,CAAC,CAAC,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC;YAClC,KAAK,IAAI,KAAK,GAAG,KAAK,CAAC;YACvB,KAAK,IAAI,CAAC,KAAK,CAAC,CAAC,CAAC,GAAG,KAAK,CAAC,IAAI,CAAC,CAAC;YACjC,WAAW,IAAI,IAAI,CAAC,GAAG,CAAC,KAAK,CAAC,CAAC;QACjC,CAAC;QAED,MAAM,EAAE,GAAG,CAAC,GAAG,CAAC,KAAK,GAAG,KAAK,CAAC,CAAC;QAC/B,MAAM,IAAI,GAAG,IAAI,CAAC,IAAI,CAAC,KAAK,GAAG,CAAC,CAAC,CAAC;QAClC,MAAM,GAAG,GAAG,WAAW,GAAG,CAAC,CAAC;QAE5B,OAAO,EAAE,EAAE,EAAE,IAAI,EAAE,GAAG,EAAE,CAAC;IAC3B,CAAC;CACF;AAED;;GAEG;AACH,MAAM,UAAU,iBAAiB,CAC/B,CAAa,EACb,CAAW,EACX,SAAiB,CAAC,EAClB,SAA8B,EAAE;IAMhC,MAAM,CAAC,GAAG,CAAC,CAAC,MAAM,CAAC;IACnB,MAAM,QAAQ,GAAG,IAAI,CAAC,KAAK,CAAC,CAAC,GAAG,MAAM,CAAC,CAAC;IACxC,MAAM,MAAM,GAAa,EAAE,CAAC;IAE5B,KAAK,IAAI,IAAI,GAAG,CAAC,EAAE,IAAI,GAAG,MAAM,EAAE,IAAI,EAAE,EAAE,CAAC;QACzC,MAAM,SAAS,GAAG,IAAI,GAAG,QAAQ,CAAC;QAClC,MAAM,OAAO,GAAG,IAAI,KAAK,MAAM,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,SAAS,GAAG,QAAQ,CAAC;QAE/D,aAAa;QACb,MAAM,MAAM,GAAG,CAAC,GAAG,CAAC,CAAC,KAAK,CAAC,CAAC,EAAE,SAAS,CAAC,EAAE,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,CAAC,CAAC,CAAC;QAC/D,MAAM,MAAM,GAAG,CAAC,GAAG,CAAC,CAAC,KAAK,CAAC,CAAC,EAAE,SAAS,CAAC,EAAE,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,CAAC,CAAC,CAAC;QAC/D,MAAM,KAAK,GAAG,CAAC,CAAC,KAAK,CAAC,SAAS,EAAE,OAAO,CAAC,CAAC;QAC1C,MAAM,KAAK,GAAG,CAAC,CAAC,KAAK,CAAC,SAAS,EAAE,OAAO,CAAC,CAAC;QAE1C,gBAAgB;QAChB,MAAM,KAAK,GAAG,IAAI,aAAa,CAAC,MAAM,CAAC,CAAC;QACxC,KAAK,CAAC,GAAG,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC;QAC1B,MAAM,KAAK,GAAG,KAAK,CAAC,KAAK,CAAC,KAAK,EAAE,KAAK,CAAC,CAAC;QACxC,MAAM,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC;IACrB,CAAC;IAED,MAAM,OAAO,GAAG,MAAM,CAAC,MAAM,CAAC,CAAC,GAAG,EAAE,CAAC,EAAE,EAAE,CAAC,GAAG,GAAG,CAAC,EAAE,CAAC,CAAC,GAAG,MAAM,CAAC;IAC/D,MAAM,QAAQ,GAAG,MAAM,CAAC,MAAM,CAAC,CAAC,GAAG,EAAE,CAAC,EAAE,EAAE,CAAC,GAAG,GAAG,CAAC,CAAC,GAAG,OAAO,CAAC,IAAI,CAAC,EAAE,CAAC,CAAC,GAAG,MAAM,CAAC;IACjF,MAAM,MAAM,GAAG,IAAI,CAAC,IAAI,CAAC,QAAQ,CAAC,CAAC;IAEnC,OAAO,EAAE,OAAO,EAAE,MAAM,EAAE,UAAU,EAAE,MAAM,EAAE,CAAC;AACjD,CAAC;AAED;;GAEG;AACH,MAAM,UAAU,OAAO,CACrB,CAAa,EACb,CAAW,EACX,SAA8B,EAAE;IAEhC,MAAM,KAAK,GAAG,IAAI,aAAa,CAAC,MAAM,CAAC,CAAC;IACxC,OAAO,KAAK,CAAC,GAAG,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC;AACzB,CAAC"}