langchain_community.embeddings.titan_takeoff
.TitanTakeoffEmbed¶
- class langchain_community.embeddings.titan_takeoff.TitanTakeoffEmbed(base_url: str = 'http://localhost', port: int = 3000, mgmt_port: int = 3001, models: List[ReaderConfig] = [])[source]¶
Interface with Takeoff Inference API for embedding models.
Use it to send embedding requests and to deploy embedding readers with Takeoff.
Examples
This is an example how to deploy an embedding model and send requests.
Initialize the Titan Takeoff embedding wrapper.
- Parameters
base_url (str, optional) – The base url where Takeoff Inference Server is
"http (listening. Defaults to) – //localhost”.
port (int, optional) – What port is Takeoff Inference API listening on.
3000. (Defaults to) –
mgmt_port (int, optional) – What port is Takeoff Management API listening on.
3001. (Defaults to) –
models (List[ReaderConfig], optional) – Any readers you’d like to spin up on.
[]. (Defaults to) –
- Raises
ImportError – If you haven’t installed takeoff-client, you will get an
ImportError. To remedy run pip install 'takeoff-client==0.4.0' –
Attributes
base_url
//localhost".
client
Takeoff Client Python SDK used to interact with Takeoff API
embed_consumer_groups
The consumer groups in Takeoff which contain embedding models
mgmt_port
The management port of the Titan Takeoff (Pro) server.
port
The port of the Titan Takeoff (Pro) server.
Methods
__init__
([base_url, port, mgmt_port, models])Initialize the Titan Takeoff embedding wrapper.
aembed_documents
(texts)Asynchronous Embed search docs.
aembed_query
(text)Asynchronous Embed query text.
embed_documents
(texts[, consumer_group])Embed documents.
embed_query
(text[, consumer_group])Embed query.
- __init__(base_url: str = 'http://localhost', port: int = 3000, mgmt_port: int = 3001, models: List[ReaderConfig] = [])[source]¶
Initialize the Titan Takeoff embedding wrapper.
- Parameters
base_url (str, optional) – The base url where Takeoff Inference Server is
"http (listening. Defaults to) – //localhost”.
port (int, optional) – What port is Takeoff Inference API listening on.
3000. (Defaults to) –
mgmt_port (int, optional) – What port is Takeoff Management API listening on.
3001. (Defaults to) –
models (List[ReaderConfig], optional) – Any readers you’d like to spin up on.
[]. (Defaults to) –
- Raises
ImportError – If you haven’t installed takeoff-client, you will get an
ImportError. To remedy run pip install 'takeoff-client==0.4.0' –
- async aembed_documents(texts: List[str]) List[List[float]] ¶
Asynchronous Embed search docs.
- Parameters
texts (List[str]) –
- Return type
List[List[float]]
- async aembed_query(text: str) List[float] ¶
Asynchronous Embed query text.
- Parameters
text (str) –
- Return type
List[float]
- embed_documents(texts: List[str], consumer_group: Optional[str] = None) List[List[float]] [source]¶
Embed documents.
- Parameters
texts (List[str]) – List of prompts/documents to embed
consumer_group (Optional[str], optional) – Consumer group to send request
None. (to containing embedding model. Defaults to) –
- Returns
List of embeddings
- Return type
List[List[float]]
- embed_query(text: str, consumer_group: Optional[str] = None) List[float] [source]¶
Embed query.
- Parameters
text (str) – Prompt/document to embed
consumer_group (Optional[str], optional) – Consumer group to send request
None. (to containing embedding model. Defaults to) –
- Returns
Embedding
- Return type
List[float]