o ZhP@snddlmZmZmZmZmZddlZddlmZddl m Z m Z ddl m Z mZdZdZGdd d e eZdS) )AnyDictListMappingOptionalN) Embeddings)get_from_dict_or_envpre_init) BaseModel ConfigDictz#sentence-transformers/clip-ViT-B-32ic@s eZdZUdZeZeed< dZe ed< dZ eed< dZ eed< d Z e eed < d Ze eed < eZeed < ed ddZededefddZedeeeffddZdeedeeefddZdeedeeefddZdedeefddZ d S)DeepInfraEmbeddingsaDeep Infra's embedding inference service. To use, you should have the environment variable ``DEEPINFRA_API_TOKEN`` set with your API token, or pass it as a named parameter to the constructor. There are multiple embeddings models available, see https://deepinfra.com/models?type=embeddings. Example: .. code-block:: python from langchain_community.embeddings import DeepInfraEmbeddings deepinfra_emb = DeepInfraEmbeddings( model_id="sentence-transformers/clip-ViT-B-32", deepinfra_api_token="my-api-key" ) r1 = deepinfra_emb.embed_documents( [ "Alpha is the first letter of Greek alphabet", "Beta is the second letter of Greek alphabet", ] ) r2 = deepinfra_emb.embed_query( "What is the second letter of Greek alphabet" ) model_idF normalizez passage: embed_instructionzquery: query_instructionN model_kwargsdeepinfra_api_token batch_sizeZforbid)extraZprotected_namespacesvaluesreturncCst|dd}||d<|S)z?Validate that api key and python package exists in environment.rZDEEPINFRA_API_TOKEN)r)clsrrrr_/var/www/html/lang_env/lib/python3.10/site-packages/langchain_community/embeddings/deepinfra.pyvalidate_environment;s z(DeepInfraEmbeddings.validate_environmentcCs d|jiS)zGet the identifying parameters.r )r selfrrr_identifying_paramsDs z'DeepInfraEmbeddings._identifying_paramsinputc Cs|jpi}d|jdd}ztjd|j|||jd|d}Wntjjy7}ztd|d}~ww|j dkrGtd |j |j fz | }|d }W|Stjj yk}z td |d |j d}~ww) Nzbearer zapplication/json) Authorizationz Content-Typez'https://api.deepinfra.com/v1/inference/)Zinputsr)headersjsonz$Error raised by inference endpoint: z/Error raised by inference API HTTP code: %s, %s embeddingszError raised by inference API: z . Response: ) rrrequestspostr r exceptionsRequestException ValueError status_codetextr!JSONDecodeError)rrZ _model_kwargsr resetr#rrr_embedIs<       zDeepInfraEmbeddings._embedtextscsTg}fdd|DfddtdtjD}|D] }||7}q|S)aQEmbed documents using a Deep Infra deployed embedding model. For larger batches, the input list of texts is chunked into smaller batches to avoid exceeding the maximum request size. Args: texts: The list of texts to embed. Returns: List of embeddings, one for each text. csg|] }j|qSr)r).0r*rrr vsz7DeepInfraEmbeddings.embed_documents..csg|] }||jqSr)r)r1iZinstruction_pairsrrrr2xsr)rangelenrr/)rr0r#chunkschunkrr4rembed_documentsis  z#DeepInfraEmbeddings.embed_documentsr*cCs"|j|}||gd}|S)zEmbed a query using a Deep Infra deployed embedding model. Args: text: The text to embed. Returns: Embeddings for the text. r)rr/)rr*Zinstruction_pairZ embeddingrrr embed_querys zDeepInfraEmbeddings.embed_query)!__name__ __module__ __qualname____doc__DEFAULT_MODEL_IDr str__annotations__rboolrrrrdictrMAX_BATCH_SIZErintr Z model_configr rrpropertyrrrrfloatr/r9r:rrrrr s0        r )typingrrrrrr$Zlangchain_core.embeddingsrZlangchain_core.utilsrr Zpydanticr r r?rDr rrrrs