gistfile1.txt

import logging
import os

from qdrant_client import QdrantClient, AsyncQdrantClient

from llama_index.vector_stores.qdrant import QdrantVectorStore
from llama_index.core.indices.vector_store.base import VectorStoreIndex
from llama_index.core.schema import Document


qdrant_client = QdrantClient(url=os.environ['QDRANT_URL'],
                             api_key=os.environ['QDRANT_API_KEY'],  # ,prefer_grpc=True,  grpc_port=30080
                             timeout=600
                             )
qdrant_async = AsyncQdrantClient(url=os.environ['QDRANT_URL'],
                                 api_key=os.environ['QDRANT_API_KEY'],  # ,prefer_grpc=True,grpc_port=30080
                                 timeout=600
                                 )
from llama_index.embeddings.langchain import LangchainEmbedding
from langchain_community.embeddings.localai import  LocalAIEmbeddings

logging.basicConfig(level=logging.DEBUG,
                    format='%(asctime)s.%(msecs)03d - %(name)s/%(levelname)s - %(message)s',
                    datefmt='%Y-%m-%d %H:%M:%S')

vector_store = QdrantVectorStore(
    "pydv9991_test",
    client=qdrant_client,
    aclient=qdrant_async,
    # prefer_grpc=True,
    # enable_hybrid=True,
    # fastembed_sparse_model=FastEmbedSparse(model_name="Qdrant/bm42-all-minilm-l6-v2-attentions"),
    batch_size=64,
)

rag_object = VectorStoreIndex.from_vector_store(
    vector_store=vector_store,
    use_async=True,
    show_progress=True,
    embed_model= LangchainEmbedding(LocalAIEmbeddings(model=os.environ['EMBEDDINGS_MODEL'],
                              openai_api_base=os.environ['EMBEDDINGS_URL'],
                              openai_api_key=os.environ['EMBEDDINGS_KEY'],))
    # quantization_config=models.ScalarQuantization(
    #     scalar=models.ScalarQuantizationConfig(
    #         type=models.ScalarType.INT8,
    #         always_ram=True,
    #     ),
    # ),
    # optimizers_config=models.OptimizersConfigDiff(default_segment_number=16),
    # hnsw_config=models.HnswConfigDiff(on_disk=False),
)

rag_object.build_index_from_nodes([Document(text="whatever "+"lorem ipsum dolor sit amet"*i) for i in range(10)])
print(rag_object.index_struct.nodes_dict)