Spaces:
Paused
Paused
File size: 4,935 Bytes
469eae6 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 |
# What is this?
## Handler file for OpenAI-like endpoints.
## Allows jina ai embedding calls - which don't allow 'encoding_format' in payload.
import json
from typing import Optional
import httpx
import litellm
from litellm.llms.custom_httpx.http_handler import (
AsyncHTTPHandler,
HTTPHandler,
get_async_httpx_client,
)
from litellm.types.utils import EmbeddingResponse
from ..common_utils import OpenAILikeBase, OpenAILikeError
class OpenAILikeEmbeddingHandler(OpenAILikeBase):
def __init__(self, **kwargs):
pass
async def aembedding(
self,
input: list,
data: dict,
model_response: EmbeddingResponse,
timeout: float,
api_key: str,
api_base: str,
logging_obj,
headers: dict,
client=None,
) -> EmbeddingResponse:
response = None
try:
if client is None or not isinstance(client, AsyncHTTPHandler):
async_client = get_async_httpx_client(
llm_provider=litellm.LlmProviders.OPENAI,
params={"timeout": timeout},
)
else:
async_client = client
try:
response = await async_client.post(
api_base,
headers=headers,
data=json.dumps(data),
) # type: ignore
response.raise_for_status()
response_json = response.json()
except httpx.HTTPStatusError as e:
raise OpenAILikeError(
status_code=e.response.status_code,
message=e.response.text if e.response else str(e),
)
except httpx.TimeoutException:
raise OpenAILikeError(
status_code=408, message="Timeout error occurred."
)
except Exception as e:
raise OpenAILikeError(status_code=500, message=str(e))
## LOGGING
logging_obj.post_call(
input=input,
api_key=api_key,
additional_args={"complete_input_dict": data},
original_response=response_json,
)
return EmbeddingResponse(**response_json)
except Exception as e:
## LOGGING
logging_obj.post_call(
input=input,
api_key=api_key,
original_response=str(e),
)
raise e
def embedding(
self,
model: str,
input: list,
timeout: float,
logging_obj,
api_key: Optional[str],
api_base: Optional[str],
optional_params: dict,
model_response: Optional[EmbeddingResponse] = None,
client=None,
aembedding=None,
custom_endpoint: Optional[bool] = None,
headers: Optional[dict] = None,
) -> EmbeddingResponse:
api_base, headers = self._validate_environment(
api_base=api_base,
api_key=api_key,
endpoint_type="embeddings",
headers=headers,
custom_endpoint=custom_endpoint,
)
model = model
data = {"model": model, "input": input, **optional_params}
## LOGGING
logging_obj.pre_call(
input=input,
api_key=api_key,
additional_args={"complete_input_dict": data, "api_base": api_base},
)
if aembedding is True:
return self.aembedding(data=data, input=input, logging_obj=logging_obj, model_response=model_response, api_base=api_base, api_key=api_key, timeout=timeout, client=client, headers=headers) # type: ignore
if client is None or isinstance(client, AsyncHTTPHandler):
self.client = HTTPHandler(timeout=timeout) # type: ignore
else:
self.client = client
## EMBEDDING CALL
try:
response = self.client.post(
api_base,
headers=headers,
data=json.dumps(data),
) # type: ignore
response.raise_for_status() # type: ignore
response_json = response.json() # type: ignore
except httpx.HTTPStatusError as e:
raise OpenAILikeError(
status_code=e.response.status_code,
message=e.response.text,
)
except httpx.TimeoutException:
raise OpenAILikeError(status_code=408, message="Timeout error occurred.")
except Exception as e:
raise OpenAILikeError(status_code=500, message=str(e))
## LOGGING
logging_obj.post_call(
input=input,
api_key=api_key,
additional_args={"complete_input_dict": data},
original_response=response_json,
)
return litellm.EmbeddingResponse(**response_json)
|