mirror of
https://github.com/danswer-ai/danswer.git
synced 2025-04-12 22:09:36 +02:00
54b883d0ca
* fix large docs selected in chat pruning * better approach to length restriction * comments * comments * fix unit tests and minor pruning bug * remove prints
99 lines
4.5 KiB
Python
99 lines
4.5 KiB
Python
import os
|
|
|
|
INPUT_PROMPT_YAML = "./onyx/seeding/input_prompts.yaml"
|
|
PROMPTS_YAML = "./onyx/seeding/prompts.yaml"
|
|
PERSONAS_YAML = "./onyx/seeding/personas.yaml"
|
|
USER_FOLDERS_YAML = "./onyx/seeding/user_folders.yaml"
|
|
NUM_RETURNED_HITS = 50
|
|
# Used for LLM filtering and reranking
|
|
# We want this to be approximately the number of results we want to show on the first page
|
|
# It cannot be too large due to cost and latency implications
|
|
NUM_POSTPROCESSED_RESULTS = 20
|
|
|
|
# May be less depending on model
|
|
MAX_CHUNKS_FED_TO_CHAT = float(os.environ.get("MAX_CHUNKS_FED_TO_CHAT") or 10.0)
|
|
# For Chat, need to keep enough space for history and other prompt pieces
|
|
# ~3k input, half for docs, half for chat history + prompts
|
|
CHAT_TARGET_CHUNK_PERCENTAGE = 512 * 3 / 3072
|
|
|
|
# Maximum percentage of the context window to fill with selected sections
|
|
SELECTED_SECTIONS_MAX_WINDOW_PERCENTAGE = 0.8
|
|
|
|
# 1 / (1 + DOC_TIME_DECAY * doc-age-in-years), set to 0 to have no decay
|
|
# Capped in Vespa at 0.5
|
|
DOC_TIME_DECAY = float(
|
|
os.environ.get("DOC_TIME_DECAY") or 0.5 # Hits limit at 2 years by default
|
|
)
|
|
BASE_RECENCY_DECAY = 0.5
|
|
FAVOR_RECENT_DECAY_MULTIPLIER = 2.0
|
|
# For the highest matching base size chunk, how many chunks above and below do we pull in by default
|
|
# Note this is not in any of the deployment configs yet
|
|
# Currently only applies to search flow not chat
|
|
CONTEXT_CHUNKS_ABOVE = int(os.environ.get("CONTEXT_CHUNKS_ABOVE") or 1)
|
|
CONTEXT_CHUNKS_BELOW = int(os.environ.get("CONTEXT_CHUNKS_BELOW") or 1)
|
|
# Whether the LLM should be used to decide if a search would help given the chat history
|
|
DISABLE_LLM_CHOOSE_SEARCH = (
|
|
os.environ.get("DISABLE_LLM_CHOOSE_SEARCH", "").lower() == "true"
|
|
)
|
|
DISABLE_LLM_QUERY_REPHRASE = (
|
|
os.environ.get("DISABLE_LLM_QUERY_REPHRASE", "").lower() == "true"
|
|
)
|
|
# 1 edit per 20 characters, currently unused due to fuzzy match being too slow
|
|
QUOTE_ALLOWED_ERROR_PERCENT = 0.05
|
|
QA_TIMEOUT = int(os.environ.get("QA_TIMEOUT") or "60") # 60 seconds
|
|
# Weighting factor between Vector and Keyword Search, 1 for completely vector search
|
|
HYBRID_ALPHA = max(0, min(1, float(os.environ.get("HYBRID_ALPHA") or 0.5)))
|
|
HYBRID_ALPHA_KEYWORD = max(
|
|
0, min(1, float(os.environ.get("HYBRID_ALPHA_KEYWORD") or 0.4))
|
|
)
|
|
# Weighting factor between Title and Content of documents during search, 1 for completely
|
|
# Title based. Default heavily favors Content because Title is also included at the top of
|
|
# Content. This is to avoid cases where the Content is very relevant but it may not be clear
|
|
# if the title is separated out. Title is most of a "boost" than a separate field.
|
|
TITLE_CONTENT_RATIO = max(
|
|
0, min(1, float(os.environ.get("TITLE_CONTENT_RATIO") or 0.10))
|
|
)
|
|
|
|
# A list of languages passed to the LLM to rephase the query
|
|
# For example "English,French,Spanish", be sure to use the "," separator
|
|
MULTILINGUAL_QUERY_EXPANSION = os.environ.get("MULTILINGUAL_QUERY_EXPANSION") or None
|
|
LANGUAGE_HINT = "\n" + (
|
|
os.environ.get("LANGUAGE_HINT")
|
|
or "IMPORTANT: Respond in the same language as my query!"
|
|
)
|
|
LANGUAGE_CHAT_NAMING_HINT = (
|
|
os.environ.get("LANGUAGE_CHAT_NAMING_HINT")
|
|
or "The name of the conversation must be in the same language as the user query."
|
|
)
|
|
|
|
# Number of prompts each persona should have
|
|
NUM_PERSONA_PROMPTS = 4
|
|
NUM_PERSONA_PROMPT_GENERATION_CHUNKS = 5
|
|
|
|
# Agentic search takes significantly more tokens and therefore has much higher cost.
|
|
# This configuration allows users to get a search-only experience with instant results
|
|
# and no involvement from the LLM.
|
|
# Additionally, some LLM providers have strict rate limits which may prohibit
|
|
# sending many API requests at once (as is done in agentic search).
|
|
# Whether the LLM should evaluate all of the document chunks passed in for usefulness
|
|
# in relation to the user query
|
|
DISABLE_LLM_DOC_RELEVANCE = (
|
|
os.environ.get("DISABLE_LLM_DOC_RELEVANCE", "").lower() == "true"
|
|
)
|
|
|
|
# Stops streaming answers back to the UI if this pattern is seen:
|
|
STOP_STREAM_PAT = os.environ.get("STOP_STREAM_PAT") or None
|
|
|
|
# Set this to "true" to hard delete chats
|
|
# This will make chats unviewable by admins after a user deletes them
|
|
# As opposed to soft deleting them, which just hides them from non-admin users
|
|
HARD_DELETE_CHATS = os.environ.get("HARD_DELETE_CHATS", "").lower() == "true"
|
|
|
|
# Internet Search
|
|
BING_API_KEY = os.environ.get("BING_API_KEY") or None
|
|
|
|
# Enable in-house model for detecting connector-based filtering in queries
|
|
ENABLE_CONNECTOR_CLASSIFIER = os.environ.get("ENABLE_CONNECTOR_CLASSIFIER", False)
|
|
|
|
VESPA_SEARCHER_THREADS = int(os.environ.get("VESPA_SEARCHER_THREADS") or 2)
|