Integrate Qdrant as another vector store (run-llama#339)

Co-authored-by: Jerry Liu <jerry@robustintelligence.com>

Author: kacperlukawski

docs/_static/vector_stores/qdrant_index_0.png

@@ -27,6 +27,7 @@ All readers can be imported through `from gpt_index.readers import ...`. A subse
 
 See [How to use Vector Stores with GPT Index](vector_stores.md) for a more thorough guide on integrating vector stores with GPT Index.
 
+- Qdrant (`QdrantReader`)
 - Weaviate (`WeaviateReader`)
 - Pinecone (`PineconeReader`)
 - Faiss (`FaissReader`)

docs/_static/vector_stores/qdrant_reader.png

@@ -9,6 +9,7 @@ GPT Index offers multiple integration points with vector stores / vector databas
 ## Loading Data from Vector Stores using Data Connector
 GPT Index supports loading data from the following sources. See [Data Connectors](data_connectors.md) for more details and API documentation.
 
+- Qdrant (`QdrantReader`) [Installation](https://qdrant.tech/documentation/install/) [Python Client](https://qdrant.tech/documentation/install/#python-client)
 - Weaviate (`WeaviateReader`). [Installation](https://weaviate.io/developers/weaviate/current/getting-started/installation.html). [Python Client](https://weaviate.io/developers/weaviate/current/client-libraries/python.html).
 - Pinecone (`PineconeReader`). [Installation/Quickstart](https://docs.pinecone.io/docs/quickstart).
 - Faiss (`FaissReader`). [Installation](https://github.com/facebookresearch/faiss/blob/main/INSTALL.md).
@@ -19,6 +20,9 @@ For instance, this is an example usage of the Pinecone data loader `PineconeRead
 
 ![](/_static/vector_stores/pinecone_reader.png)
 
+Qdrant stores both documents and vectors, so there is no need to provide `id_to_text_map`. This is an example of how to use it:
+
+![](/_static/vector_stores/qdrant_reader.png)
 
 NOTE: Since Weaviate can store a hybrid of document and vector objects, the user may either choose to explicitly specify `class_name` and `properties` in order to query documents, or they may choose to specify a raw GraphQL query. See below for usage.
 
@@ -37,6 +41,7 @@ These are found in the following classes:
 - `GPTFaissIndex`
 - `GPTWeaviateIndex`
 - `GPTPineconeIndex`
+- `GPTQdrantIndex`
 
 Similar to any other index within GPT Index (tree, keyword table, list), this index can be constructed upon any collection
 of documents. We use the vector store within the index to store embeddings for the input text chunks.
@@ -56,4 +61,7 @@ Once constructed, the index can be used for querying.
 **Pinecone Index Construction/Querying**
 ![](/_static/vector_stores/pinecone_index_0.png)
 
+**Qdrant Index Construction/Querying**
+![](/_static/vector_stores/qdrant_index_0.png)
+
 [Example notebooks can be found here](https://github.com/jerryjliu/gpt_index/tree/main/examples/vector_indices).

docs/how_to/data_connectors.md

@@ -0,0 +1,118 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "f3ca56f0-6ef1-426f-bac5-fd7c374d0f51",
+   "metadata": {},
+   "source": [
+    "# Qdrant Demo"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "262f990a-79c8-413a-9f3c-cd9a3c191307",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from gpt_index.readers.qdrant import QdrantReader"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "252f8163-7297-44b6-a838-709e9662f3d6",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "reader = QdrantReader(host=\"localhost\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "53b49187-8477-436c-9718-5d2f8cc6fad0",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# the query_vector is an embedding representation of your query_vector\n",
+    "# Example query vector:\n",
+    "#   query_vector=[0.3, 0.3, 0.3, 0.3, 0.3, 0.3, 0.3, 0.3]\n",
+    "\n",
+    "query_vector=[n1, n2, n3, ...]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "a88be1c4-603f-48b9-ac64-10a219af4951",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# NOTE: Required args are collection_name, query_vector.\n",
+    "# See the Python client: https://github.com/qdrant/qdrant_client\n",
+    "# for more details. \n",
+    "documents = reader.load_data(collection_name=\"demo\", query_vector=query_vector, limit=5)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "169b4273-eb20-4d06-9ffe-71320f4570f6",
+   "metadata": {},
+   "source": [
+    "### Create index"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "ac4563a1",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "index = GPTListIndex(documents)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "f06b02db",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "response = index.query(\"<query_text>\", verbose=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "97d1ae80",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "display(Markdown(f\"<b>{response}</b>\"))"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.16"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}