wehub-resource-sync
54 lines
1.7 KiB
Python
54 lines
1.7 KiB
Python
import os
|
|
|
|
from bfcl_eval.model_handler.api_inference.openai_completion import (
|
|
OpenAICompletionsHandler,
|
|
)
|
|
from bfcl_eval.constants.enums import ModelStyle
|
|
from bfcl_eval.model_handler.utils import (
|
|
combine_consecutive_user_prompts,
|
|
default_decode_ast_prompting,
|
|
default_decode_execute_prompting,
|
|
system_prompt_pre_processing_chat_model,
|
|
)
|
|
from openai import OpenAI
|
|
|
|
|
|
class NvidiaHandler(OpenAICompletionsHandler):
|
|
def __init__(
|
|
self,
|
|
model_name,
|
|
temperature,
|
|
registry_name,
|
|
is_fc_model,
|
|
**kwargs,
|
|
) -> None:
|
|
super().__init__(model_name, temperature, registry_name, is_fc_model, **kwargs)
|
|
self.model_style = ModelStyle.OPENAI_COMPLETIONS
|
|
self.client = OpenAI(
|
|
base_url="https://integrate.api.nvidia.com/v1",
|
|
api_key=os.getenv("NVIDIA_API_KEY"),
|
|
)
|
|
|
|
def decode_ast(self, result, language, has_tool_call_tag):
|
|
return default_decode_ast_prompting(result, language, has_tool_call_tag)
|
|
|
|
def decode_execute(self, result, has_tool_call_tag):
|
|
return default_decode_execute_prompting(result, has_tool_call_tag)
|
|
|
|
#### Prompting methods ####
|
|
|
|
def _pre_query_processing_prompting(self, test_entry: dict) -> dict:
|
|
functions: list = test_entry["function"]
|
|
test_entry_id: str = test_entry["id"]
|
|
|
|
test_entry["question"][0] = system_prompt_pre_processing_chat_model(
|
|
test_entry["question"][0], functions, test_entry_id
|
|
)
|
|
|
|
for round_idx in range(len(test_entry["question"])):
|
|
test_entry["question"][round_idx] = combine_consecutive_user_prompts(
|
|
test_entry["question"][round_idx]
|
|
)
|
|
|
|
return {"message": []}
|