dsrag-sdk 0.0.5

State-of-the-art RAG pipeline from D-Star AI

Overview

This python package is only useful for a custom cloud deployment of dsRAG

Usage

pip install dsrag-sdk

Setup

You can set the base url that the sdk connects to by running the following:

from dsrag_sdk.main import set_base_url
set_base_url("http://0.0.0.0:8000")

Knowledge bases

from dsrag_sdk.knowledge_bases import create_knowledge_base, list_knowledge_bases, get_knowledge_base, delete_knowledge_base, update_knowledge_base

Create knowledge base

kb = create_knowledge_base(title="sdk_test_kb", supp_id="test-supp-id")

List knowledge bases

knowledge_bases = list_knowledge_bases()

List with supp_id filter

knowledge_bases = list_knowledge_bases(supp_id="test-supp-id")

Get a knowledge base by id

knowledge_base = get_knowledge_base(knowledge_base_id="test-kb-id")

Update a knowledge base

knowledge_base = update_knowledge_base(knowledge_base_id="test-kb-id", title="updated-title", supp_id="updated-supp-id")

Delete a knowledge base

response = delete_knowledge_base(knowledge_base_id="test-kb-id")

Documents

from dsrag_sdk.documents import create_document_via_text, list_documents, get_document, delete_document

Add a document via text

kb_id = "test-kb-id"
document_text = "This is some test text for document upload"
document = create_document_via_text(kb_id, document_text)

List documents

documents = list_documents(kb_id)

List documents with supp_id filter

documents = list_documents(knowledge_base_id="test-kb-id", supp_id="test-supp-id")

Get document by id

document = get_document(knowledge_base_id="test-kb-id", document_id="test-document-id")

Delete a document

response = delete_document(knowledge_base_id="test-kb-id", document_id="test-document-id")

Chat

from dsrag_sdk.chat import create_chat_thread, list_chat_threads, update_chat_thread, get_chat_thread, delete_chat_thread, get_chat_response, list_thread_interactions

Create a chat thread

response = create_chat_thread(knowledge_base_ids=[kb_id])

List chat threads

chat_threads = list_chat_threads()

Filter with supp_id

chat_threads = list_chat_threads(supp_id="test-supp-id")

Get chat thread by id

chat_thread = get_chat_threads(thread_id="test-thread-id")

Get a chat response

response = get_chat_response(thread_id="test-thread-id", input="What is AGI?")

Update a chat thread

chat_thread = update_chat_thread(thread_id="test-thread-id", knowledge_base_ids=[kb_id_1, kn_id_2], supp_id="updated-supp-id", model="gpt-4o-mini")

Delete a chat thread

response = delete_chat_thread(thread_id="test-thread-id")