generative-bayesian-network 0.1.0-beta.1 → 2.0.0-dev.0

package/{LICENSE → LICENSE.md}

@@ -178,7 +178,7 @@
    APPENDIX: How to apply the Apache License to your work.
 
       To apply the Apache License to your work, attach the following
-      boilerplate notice, with the fields enclosed by brackets "[]"
+      boilerplate notice, with the fields enclosed by brackets "{}"
       replaced with your own identifying information. (Don't include
       the brackets!)  The text should be enclosed in the appropriate
       comment syntax for the file format. We also recommend that a
@@ -186,7 +186,7 @@
       same "printed page" as the copyright notice for easier
       identification within third-party archives.
 
-   Copyright [yyyy] [name of copyright owner]
+   Copyright 2018 Apify Technologies s.r.o.
 
    Licensed under the Apache License, Version 2.0 (the "License");
    you may not use this file except in compliance with the License.

package/README.md

@@ -1,76 +1,3 @@
-# Generative bayesian network
-NodeJs package containing a bayesian network capable of randomly sampling from a distribution defined by a json file.
-
-<!-- toc -->
-
-- [Installation](#installation)
-- [Usage](#usage)
-- [Examples](#examples)
-- [API Reference](#api-reference)
-
-<!-- tocstop -->
-
-## Installation
-
-## Usage
-
-## Examples
-
-## API Reference
-All public classes, methods and their parameters can be inspected in this API reference.
-
-<a name="BayesianNetwork"></a>
-
-### BayesianNetwork
-BayesianNetwork is an implementation of a bayesian network capable of randomly sampling from the distribution
-represented by the network.
-
-
-* [BayesianNetwork](#BayesianNetwork)
-    * [`new BayesianNetwork(networkDefinition)`](#new_BayesianNetwork_new)
-    * [`.generateSample(inputValues)`](#BayesianNetwork+generateSample)
-    * [`.generateConsistentSampleWhenPossible(valuePossibilities)`](#BayesianNetwork+generateConsistentSampleWhenPossible)
-
-
-* * *
-
-<a name="new_BayesianNetwork_new"></a>
-
-#### `new BayesianNetwork(networkDefinition)`
-
-| Param | Type | Description |
-| --- | --- | --- |
-| networkDefinition | <code>object</code> | object defining the network structure and distributions |
-
-
-* * *
-
-<a name="BayesianNetwork+generateSample"></a>
-
-#### `bayesianNetwork.generateSample(inputValues)`
-Randomly samples from the distribution represented by the bayesian network.
-Can generate samples not found in the data used to create the definition file.
-
-
-| Param | Type | Description |
-| --- | --- | --- |
-| inputValues | <code>object</code> | node values that are known already |
-
-
-* * *
-
-<a name="BayesianNetwork+generateConsistentSampleWhenPossible"></a>
-
-#### `bayesianNetwork.generateConsistentSampleWhenPossible(valuePossibilities)`
-Randomly samples from the distribution represented by the bayesian network.
-Cannot generate samples not found in the data used to create the definition file,
-so it only generates samples when it is possible to be consistent with the data.
-
-
-| Param | Type | Description |
-| --- | --- | --- |
-| valuePossibilities | <code>object</code> | a dictionary of lists of possible values for nodes                                      (if a node isn't present in the dictionary, all values are possible) |
-
-
-* * *
+# Fingerprint suite
 
+This repository contains a set of fingerprinting tools developed by Apify.

package/index.js

@@ -0,0 +1,6 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.BayesianNetwork = void 0;
+var bayesian_network_1 = require("./bayesian-network");
+Object.defineProperty(exports, "BayesianNetwork", { enumerable: true, get: function () { return bayesian_network_1.BayesianNetwork; } });
+//# sourceMappingURL=index.js.map

package/package.json

@@ -1,37 +1,40 @@
 {
-  "author": {
-    "name": "Apify"
-  },
-  "files": ["src"],
-  "bugs": {
-    "url": "https://github.com/apify/generative-bayesian-network/issues"
-  },
-  "dependencies": {
-    "ow": "^0.23.0"
-  },
-  "description": "An implementation of a bayesian network usable for .",
-  "devDependencies": {
-    "@apify/eslint-config": "^0.1.3",
-    "eslint": "^7.19.0",
-    "fs-extra": "^9.1.0",
-    "jest": "^26.6.3",
-    "jsdoc-to-markdown": "^7.0.0",
-    "markdown-toc": "^1.2.0"
-  },
-  "homepage": "https://github.com/apify/generative-bayesian-network#readme",
-  "license": "Apache-2.0",
-  "main": "src/main.js",
-  "name": "generative-bayesian-network",
-  "repository": {
-    "type": "git",
-    "url": "git+https://github.com/apify/generative-bayesian-network.git"
-  },
-  "scripts": {
-    "build-docs": "npm run build-toc && node docs/build-docs.js",
-    "build-toc": "markdown-toc docs/README.md -i",
-    "lint": "eslint ./src --ext .js,.jsx",
-    "lint:fix": "eslint ./src --ext .js,.jsx --fix",
-    "test": "jest --maxWorkers=3 --forceExit"
-  },
-  "version": "0.1.0-beta.1"
+    "name": "generative-bayesian-network",
+    "version": "2.0.0-dev.0",
+    "author": {
+        "name": "Apify"
+    },
+    "files": [
+        "dist"
+    ],
+    "main": "index.js",
+    "module": "index.mjs",
+    "types": "index.d.ts",
+    "exports": {
+        ".": {
+            "import": "./index.mjs",
+            "require": "./index.js"
+        }
+    },
+    "bugs": {
+        "url": "https://github.com/apify/generative-bayesian-network/issues"
+    },
+    "dependencies": {
+        "adm-zip": "^0.5.9",
+        "danfojs-node": "^1.1.1"
+    },
+    "description": "An implementation of a bayesian network usable for .",
+    "homepage": "https://github.com/apify/generative-bayesian-network#readme",
+    "license": "Apache-2.0",
+    "repository": {
+        "type": "git",
+        "url": "git+https://github.com/apify/generative-bayesian-network.git"
+    },
+    "scripts": {
+        "build": "npm run clean && npm run compile",
+        "clean": "rimraf ./dist",
+        "compile": "tsc -p tsconfig.build.json && gen-esm-wrapper ./index.js ./index.mjs",
+        "copy": "ts-node -T ../../scripts/copy.ts"
+    },
+    "gitHead": "2dd3d8c9ced1a14cb74ef16ebdd50b349bb5971b"
 }

package/src/bayesian-network.js

@@ -1,85 +0,0 @@
-const { BayesianNode } = require('./bayesian-node');
-
-/**
- * BayesianNetwork is an implementation of a bayesian network capable of randomly sampling from the distribution
- * represented by the network.
- */
-class BayesianNetwork {
-    /**
-     * @param {object} networkDefinition - object defining the network structure and distributions
-     */
-    constructor(networkDefinition) {
-        this.nodesInSamplingOrder = networkDefinition.nodes.map((nodeDefinition) => new BayesianNode(nodeDefinition));
-        this.nodesByName = {};
-        for (const node of this.nodesInSamplingOrder) {
-            this.nodesByName[node.name] = node;
-        }
-    }
-
-    /**
-     * Randomly samples from the distribution represented by the bayesian network.
-     * Can generate samples not found in the data used to create the definition file.
-     * @param {object} inputValues - node values that are known already
-     */
-    generateSample(inputValues = {}) {
-        const sample = inputValues;
-        for (const node of this.nodesInSamplingOrder) {
-            if (!(node.name in sample)) {
-                sample[node.name] = node.sample(sample);
-            }
-        }
-
-        return sample;
-    }
-
-    /**
-     * Randomly samples from the distribution represented by the bayesian network.
-     * Cannot generate samples not found in the data used to create the definition file,
-     * so it only generates samples when it is possible to be consistent with the data.
-     * @param {object} valuePossibilities - a dictionary of lists of possible values for nodes
-     *                                      (if a node isn't present in the dictionary, all values are possible)
-     */
-    generateConsistentSampleWhenPossible(valuePossibilities) {
-        return this._recursivelyGenerateConsistentSampleWhenPossible({}, valuePossibilities, 0);
-    }
-
-    /**
-     * Randomly samples from the distribution represented by the bayesian network
-     * @param {object} sampleSoFar - node values that are known already
-     * @param {object} valuePossibilities - a dictionary of lists of possible values for nodes
-     *                                      (if a node isn't present in the dictionary, all values are possible)
-     * @param {number} depth - in what depth of the recursion this function call is,
-     *                         specifies what node this function call is sampling
-     * @private
-     */
-    _recursivelyGenerateConsistentSampleWhenPossible(sampleSoFar, valuePossibilities, depth) {
-        const bannedValues = [];
-        const node = this.nodesInSamplingOrder[depth];
-
-        let sampleValue;
-        do {
-            sampleValue = node.sampleAccordingToRestrictions(sampleSoFar, valuePossibilities[node.name], bannedValues);
-
-            if (!sampleValue) break;
-
-            sampleSoFar[node.name] = sampleValue;
-
-            if (depth + 1 < this.nodesInSamplingOrder.length) {
-                const sample = this._recursivelyGenerateConsistentSampleWhenPossible(sampleSoFar, valuePossibilities, depth + 1);
-                if (sample) {
-                    return sample;
-                }
-            } else {
-                return sampleSoFar;
-            }
-
-            bannedValues.push(sampleValue);
-        } while (sampleValue);
-
-        return false;
-    }
-}
-
-module.exports = {
-    BayesianNetwork,
-};

package/src/bayesian-node.js

@@ -1,103 +0,0 @@
-/**
- * BayesianNode is an implementation of a single node in a bayesian network allowing
- * sampling from its conditional distribution.
- */
-class BayesianNode {
-    /**
-     * @param {object} nodeDefinition - node structure and distributions definition
-     *                                  taken from the network definition file
-     */
-    constructor(nodeDefinition) {
-        this.nodeDefinition = nodeDefinition;
-    }
-
-    /**
-     * Extracts unconditional probabilities of node values given the values of the parent nodes
-     * @param {object} parentValues - values of the parent nodes
-     * @private
-     */
-    _getProbabilitiesGivenKnownValues(parentValues = {}) {
-        let probabilities = this.nodeDefinition.conditionalProbabilities;
-
-        for (const parentName of this.parentNames) {
-            const parentValue = parentValues[parentName];
-            if (parentValue in probabilities.deeper) {
-                probabilities = probabilities.deeper[parentValue];
-            } else {
-                probabilities = probabilities.skip;
-            }
-        }
-        return probabilities;
-    }
-
-    /**
-     * Randomly samples from the given values using the given probabilities
-     * @param {array} possibleValues - a list of values to sample from
-     * @param {number} totalProbabilityOfPossibleValues - a sum of probabilities of possibleValues
-     *                                                    in the conditional distribution
-     * @param {object} probabilities - a dictionary of probabilities from the conditional distribution
-     *                                 indexed by the values
-     * @private
-     */
-    _sampleRandomValueFromPossibilities(possibleValues, totalProbabilityOfPossibleValues, probabilities) {
-        let chosenValue = possibleValues[0];
-        const anchor = Math.random() * totalProbabilityOfPossibleValues;
-        let cumulativeProbability = 0;
-        for (const possibleValue of possibleValues) {
-            cumulativeProbability += probabilities[possibleValue];
-            if (cumulativeProbability > anchor) {
-                chosenValue = possibleValue;
-                break;
-            }
-        }
-
-        return chosenValue;
-    }
-
-    /**
-     * Randomly samples from the conditional distribution of this node given values of parents
-     * @param {object} parentValues - values of the parent nodes
-     */
-    sample(parentValues = {}) {
-        const probabilities = this._getProbabilitiesGivenKnownValues(parentValues);
-        const possibleValues = Object.keys(probabilities);
-
-        return this._sampleRandomValueFromPossibilities(possibleValues, 1.0, probabilities);
-    }
-
-    /**
-     * Randomly samples from the conditional distribution of this node given restrictions on the possible
-     * values and the values of the parents.
-     * @param {object} parentValues - values of the parent nodes
-     * @param {object} valuePossibilities - a list of possible values for this node
-     * @param {object} bannedValues - what values of this node are banned
-     */
-    sampleAccordingToRestrictions(parentValues, valuePossibilities, bannedValues) {
-        const probabilities = this._getProbabilitiesGivenKnownValues(parentValues);
-        let totalProbability = 0.0;
-        const validValues = [];
-        const valuesInDistribution = Object.keys(probabilities);
-        const possibleValues = valuePossibilities || valuesInDistribution;
-        for (const value of possibleValues) {
-            if (!bannedValues.includes(value) && valuesInDistribution.includes(value)) {
-                validValues.push(value);
-                totalProbability += probabilities[value];
-            }
-        }
-
-        if (validValues.length === 0) return false;
-        return this._sampleRandomValueFromPossibilities(validValues, totalProbability, probabilities);
-    }
-
-    get name() {
-        return this.nodeDefinition.name;
-    }
-
-    get parentNames() {
-        return this.nodeDefinition.parentNames;
-    }
-}
-
-module.exports = {
-    BayesianNode,
-};

package/src/main.js

@@ -1,3 +0,0 @@
-const BayesianNetwork = require('./bayesian-network');
-
-module.exports = BayesianNetwork;