Spaces:
Running
Running
File size: 1,849 Bytes
9a9ffe5 1e91476 ee77d0d 1e91476 9a9ffe5 1e91476 9a9ffe5 1e91476 f755aca 9a9ffe5 1e91476 ee77d0d 1e91476 ee77d0d 1e91476 |
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 |
from .input_processing import parse_app_request, initialize_conversation, parse_prompt, check_and_modify_messages
from .response_generation import generate_sim
from .response_processing import process_model_response
def process_app_request(app_request: dict, endpoint_url: str, endpoint_bearer_token: str) -> dict:
"""Process the app request and return the response in the required format."""
############################# Input Processing ###################################
# Parse the app request into model_input and extract the prompt
model_input, prompt, conversation_id = parse_app_request(app_request)
# Initialize the conversation (adds the system message)
model_input = initialize_conversation(model_input, conversation_id)
# Parse the prompt into messages
prompt_messages = parse_prompt(prompt)
# Check and modify messages adding a suicidal statement to avoid model going out of character
prompt_messages = check_and_modify_messages(prompt_messages)
# Append the messages parsed from the app prompt to the conversation history
model_input['messages'].extend(prompt_messages)
####################################################################################
####################### Output Generation & Processing #############################
# Generate the assistant's response (texter's reply)
completion = generate_sim(model_input, endpoint_url, endpoint_bearer_token)
# Process the raw model response (parse, guardrails, split)
final_response = process_model_response(completion, model_input, endpoint_url, endpoint_bearer_token)
# Format the response for the APP
response = {"predictions": [{"generated_text": final_response}]}
####################################################################################
return response |