Chroma 读取器¶

如果你在 colab 上打开此 Notebook，你可能需要安装 LlamaIndex 🦙。

In [ ]

%pip install llama-index-readers-chroma

%pip install llama-index-readers-chroma

In [ ]

!pip install llama-index

!pip install llama-index

In [ ]

import logging
import sys

logging.basicConfig(stream=sys.stdout, level=logging.INFO)
logging.getLogger().addHandler(logging.StreamHandler(stream=sys.stdout))

import logging import sys logging.basicConfig(stream=sys.stdout, level=logging.INFO) logging.getLogger().addHandler(logging.StreamHandler(stream=sys.stdout))

In [ ]

from llama_index.readers.chroma import ChromaReader

from llama_index.readers.chroma import ChromaReader

In [ ]

# The chroma reader loads data from a persisted Chroma collection.
# This requires a collection name and a persist directory.

reader = ChromaReader(
    collection_name="chroma_collection",
    persist_directory="examples/data_connectors/chroma_collection",
)

# chroma 读取器从持久化的 Chroma collection 加载数据。 # 这需要一个 collection 名称和一个持久化目录。 reader = ChromaReader( collection_name="chroma_collection", persist_directory="examples/data_connectors/chroma_collection", )

In [ ]

# the query_vector is an embedding representation of your query.
# Example query vector:
#   query_vector=[0.3, 0.3, 0.3, 0.3, 0.3, 0.3, 0.3, 0.3]

query_vector = [n1, n2, n3, ...]

# query_vector 是你的查询的嵌入表示。 # 示例查询向量： # query_vector=[0.3, 0.3, 0.3, 0.3, 0.3, 0.3, 0.3, 0.3] query_vector = [n1, n2, n3, ...]

In [ ]

# NOTE: Required args are collection_name, query_vector.
# See the Python client: https://github.com/chroma-core/chroma
# for more details.
documents = reader.load_data(
    collection_name="demo", query_vector=query_vector, limit=5
)

# 注意：必需参数是 collection_name, query_vector。 # 更多详情请参阅 Python 客户端：https://github.com/chroma-core/chroma # for more details. documents = reader.load_data( collection_name="demo", query_vector=query_vector, limit=5 )

创建索引¶

In [ ]

from llama_index.core import SummaryIndex

index = SummaryIndex.from_documents(documents)

from llama_index.core import SummaryIndex index = SummaryIndex.from_documents(documents)

In [ ]

# set Logging to DEBUG for more detailed outputs
query_engine = index.as_query_engine()
response = query_engine.query("<query_text>")

# 设置日志级别为 DEBUG 以获取更详细的输出 query_engine = index.as_query_engine() response = query_engine.query("")

In [ ]

display(Markdown(f"<b>{response}</b>"))

display(Markdown(f"{response}"))