vectra 0.1.0 → 0.1.2

package/README.md

@@ -1,10 +1,95 @@
-# vectra
+# Vectra
 Vectra is a local vector database for Node.js with features similar to [Pinecone](https://www.pinecone.io/) or [Qdrant](https://qdrant.tech/) but built using local files. Each Vectra index is a folder on disk. There's an `index.json` file in the folder that contains all the vectors for the index along with any indexed metadata.  When you create an index you can specify which metadata properties to index and only those fields will be stored in the `index.json` file. All of the other metadata for an item will be stored on disk in a separate file keyed by a GUID.
 
 When queryng Vectra you'll be able to use the same subset of [Mongo DB query operators](https://www.mongodb.com/docs/manual/reference/operator/query/) that Pinecone supports and the results will be returned sorted by simularity. Every item in the index will first be filtered by metadata and then ranked for simularity. Even though every item is evaluated its all in memory so it should by nearly instantanious. Likely 1ms - 2ms for even a rather large index. Smaller indexes should be <1ms.
 
-Keep in mind that your entire Vectra index is loaded into memory so it's not well suited for scenarios like long term chat bot memory. Use a really vector DB for that. Vectra is intended to be used in scenarios where you have a small corpus of mostly static data you'd like to include in your prompt. Infinite few shot examples would be a great use case for Vectra or even just a single document you want to ask questions over.
+Keep in mind that your entire Vectra index is loaded into memory so it's not well suited for scenarios like long term chat bot memory. Use a real vector DB for that. Vectra is intended to be used in scenarios where you have a small corpus of mostly static data that you'd like to include in your prompt. Infinite few shot examples would be a great use case for Vectra or even just a single document you want to ask questions over.
 
 Pinecone style namespaces aren't directly supported but you could easily mimic them by creating a separate Vectra index (and folder) for each namespace.
 
-Sample code and the actual code coming the week. GPT and I are still writting it :)
+## Installation
+
+```
+$ npm install vectra
+```
+
+## Usage
+
+First create an instance of `LocalIndex` with the path to the folder where you want you're items stored:
+
+```typescript
+import { LocalIndex } from 'vectra';
+
+const index = new LocalIndex(path.join(__dirname, '..', 'index'));
+```
+
+Next, from inside an async function, create your index:
+
+```typescript
+if (!await index.isIndexCreated()) {
+    await index.createIndex();
+}
+```
+
+Add some items to your index:
+
+```typescript
+import { OpenAIApi, Configuration } from 'openai';
+
+const configuration = new Configuration({
+    apiKey: `<YOUR_KEY>`,
+});
+
+const api = new OpenAIApi(configuration);
+
+async function getVector(text: string) {
+    const response = await api.createEmbedding({
+        'model': 'text-embedding-ada-002',
+        'input': text,
+    });
+    return response.data.data[0].embedding;
+}
+
+async function addItem(text: string) {
+    await index.insertItem({
+        vector: await getVector(text),
+        metadata: { text }
+    });
+}
+
+// Add items
+await addItem('apple');
+await addItem('oranges');
+await addItem('red');
+await addItem('blue');
+```
+
+Then query for items:
+
+```typescript
+async function query(text: string) {
+    const vector = await getVector(input);
+    const results = await index.queryItems(vector, 3);
+    if (results.length > 0) {
+        for (const result of results) {
+            console.log(`[${result.score}] ${result.item.metadata.text}`);
+        }
+    } else {
+        console.log(`No results found.`);
+    }
+}
+
+await query('green');
+/*
+[0.9036569942401076] blue
+[0.8758153664568566] red
+[0.8323828606103998] apple
+*/
+
+await query('banana');
+/*
+[0.9033128691220631] apple
+[0.8493374123092652] oranges
+[0.8415324469533297] blue
+*/
+```

package/package.json

@@ -2,7 +2,7 @@
     "name": "vectra",
     "author": "Steven Ickman",
     "description": "A vector database that uses the local file system for storage.",
-    "version": "0.1.0",
+    "version": "0.1.2",
     "license": "MIT",
     "keywords": [
         "gpt"

package/lib/utilities.d.ts

@@ -1,5 +0,0 @@
-export declare function dotProduct(arr1: number[], arr2: number[]): number;
-export declare function normalize(arr: number[]): number;
-export declare function cosineSimilarity(arr1: number[], arr2: number[]): number;
-export declare function normalizedCosineSimilarity(arr1: number[], norm1: number, arr2: number[], norm2: number): number;
-//# sourceMappingURL=utilities.d.ts.map

package/lib/utilities.d.ts.map

@@ -1 +0,0 @@
-{"version":3,"file":"utilities.d.ts","sourceRoot":"","sources":["../src/utilities.ts"],"names":[],"mappings":"AACA,wBAAgB,UAAU,CAAC,IAAI,EAAE,MAAM,EAAE,EAAE,IAAI,EAAE,MAAM,EAAE,UAUxD;AAED,wBAAgB,SAAS,CAAC,GAAG,EAAE,MAAM,EAAE,UAUtC;AAED,wBAAgB,gBAAgB,CAAC,IAAI,EAAE,MAAM,EAAE,EAAE,IAAI,EAAE,MAAM,EAAE,UAG9D;AAED,wBAAgB,0BAA0B,CAAC,IAAI,EAAE,MAAM,EAAE,EAAE,KAAK,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,EAAE,EAAE,KAAK,EAAE,MAAM,UAGtG"}

package/lib/utilities.js

@@ -1,38 +0,0 @@
-"use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.normalizedCosineSimilarity = exports.cosineSimilarity = exports.normalize = exports.dotProduct = void 0;
-function dotProduct(arr1, arr2) {
-    // Initialize a variable to store the sum of the products
-    let sum = 0;
-    // Loop through the elements of the arrays
-    for (let i = 0; i < arr1.length; i++) {
-        // Multiply the corresponding elements and add them to the sum
-        sum += arr1[i] * arr2[i];
-    }
-    // Return the sum
-    return sum;
-}
-exports.dotProduct = dotProduct;
-function normalize(arr) {
-    // Initialize a variable to store the sum of the squares
-    let sum = 0;
-    // Loop through the elements of the array
-    for (let i = 0; i < arr.length; i++) {
-        // Square the element and add it to the sum
-        sum += arr[i] * arr[i];
-    }
-    // Return the square root of the sum
-    return Math.sqrt(sum);
-}
-exports.normalize = normalize;
-function cosineSimilarity(arr1, arr2) {
-    // Return the quotient of the dot product and the product of the norms
-    return dotProduct(arr1, arr2) / (normalize(arr1) * normalize(arr2));
-}
-exports.cosineSimilarity = cosineSimilarity;
-function normalizedCosineSimilarity(arr1, norm1, arr2, norm2) {
-    // Return the quotient of the dot product and the product of the norms
-    return dotProduct(arr1, arr2) / (norm1 * norm2);
-}
-exports.normalizedCosineSimilarity = normalizedCosineSimilarity;
-//# sourceMappingURL=utilities.js.map

package/lib/utilities.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"utilities.js","sourceRoot":"","sources":["../src/utilities.ts"],"names":[],"mappings":";;;AACA,SAAgB,UAAU,CAAC,IAAc,EAAE,IAAc;IACrD,yDAAyD;IACzD,IAAI,GAAG,GAAG,CAAC,CAAC;IACZ,0CAA0C;IAC1C,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,IAAI,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE;QAClC,8DAA8D;QAC9D,GAAG,IAAI,IAAI,CAAC,CAAC,CAAC,GAAG,IAAI,CAAC,CAAC,CAAC,CAAC;KAC5B;IACD,iBAAiB;IACjB,OAAO,GAAG,CAAC;AACf,CAAC;AAVD,gCAUC;AAED,SAAgB,SAAS,CAAC,GAAa;IACnC,wDAAwD;IACxD,IAAI,GAAG,GAAG,CAAC,CAAC;IACZ,yCAAyC;IACzC,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,GAAG,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE;QACjC,2CAA2C;QAC3C,GAAG,IAAI,GAAG,CAAC,CAAC,CAAC,GAAG,GAAG,CAAC,CAAC,CAAC,CAAC;KAC1B;IACD,oCAAoC;IACpC,OAAO,IAAI,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC;AAC1B,CAAC;AAVD,8BAUC;AAED,SAAgB,gBAAgB,CAAC,IAAc,EAAE,IAAc;IAC3D,sEAAsE;IACtE,OAAO,UAAU,CAAC,IAAI,EAAE,IAAI,CAAC,GAAG,CAAC,SAAS,CAAC,IAAI,CAAC,GAAG,SAAS,CAAC,IAAI,CAAC,CAAC,CAAC;AACxE,CAAC;AAHD,4CAGC;AAED,SAAgB,0BAA0B,CAAC,IAAc,EAAE,KAAa,EAAE,IAAc,EAAE,KAAa;IACnG,sEAAsE;IACtE,OAAO,UAAU,CAAC,IAAI,EAAE,IAAI,CAAC,GAAG,CAAC,KAAK,GAAG,KAAK,CAAC,CAAC;AACpD,CAAC;AAHD,gEAGC"}