sij/sijapi
1
1
Fork 0
Code Issues 3 Pull requests Projects Releases Packages Wiki Activity Actions

Auto-update: Thu Nov 14 13:07:52 PST 2024

Browse source
This commit is contained in:
sanj 2024-11-14 13:07:52 -08:00
parent 3e33c5914f
commit 3ffbe93764
1 changed files with 17 additions and 17 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

34
sijapi/routers/llm.py
View file

@ -26,7 +26,7 @@ import tempfile
import shutil import shutil
import html2text import html2text
import markdown import markdown
from sijapi import Llm, LLM_SYS_MSG, REQUESTS_DIR, OBSIDIAN_CHROMADB_COLLECTION, OBSIDIAN_VAULT_DIR, DOC_DIR, SUMMARY_INSTRUCT, SUMMARY_CHUNK_SIZE, SUMMARY_TPW, SUMMARY_CHUNK_OVERLAP, SUMMARY_LENGTH_RATIO, SUMMARY_TOKEN_LIMIT, SUMMARY_MIN_LENGTH, SUMMARY_MODEL from sijapi import Llm, LLM_SYS_MSG, REQUESTS_DIR, OBSIDIAN_CHROMADB_COLLECTION, OBSIDIAN_VAULT_DIR, DOC_DIR
from sijapi.utilities import convert_to_unix_time, sanitize_filename, ocr_pdf, clean_text, should_use_ocr, extract_text_from_pdf, extract_text_from_docx, read_text_file, str_to_bool, get_extension from sijapi.utilities import convert_to_unix_time, sanitize_filename, ocr_pdf, clean_text, should_use_ocr, extract_text_from_pdf, extract_text_from_docx, read_text_file, str_to_bool, get_extension
from sijapi.routers import tts from sijapi.routers import tts
from sijapi.routers.asr import transcribe_audio from sijapi.routers.asr import transcribe_audio
@ -512,12 +512,12 @@ def gpt4v(image_base64, prompt_sys: str, prompt_usr: str, max_tokens: int = 150)
@llm.get("/summarize") @llm.get("/summarize")
async def summarize_get(text: str = Form(None), instruction: str = Form(SUMMARY_INSTRUCT)): async def summarize_get(text: str = Form(None), instruction: str = Form(Llm.summary.instruct)):
summarized_text = await summarize_text(text, instruction) summarized_text = await summarize_text(text, instruction)
return summarized_text return summarized_text
@llm.post("/summarize") @llm.post("/summarize")
async def summarize_post(file: Optional[UploadFile] = File(None), text: Optional[str] = Form(None), instruction: str = Form(SUMMARY_INSTRUCT)): async def summarize_post(file: Optional[UploadFile] = File(None), text: Optional[str] = Form(None), instruction: str = Form(Llm.summary.instruct)):
text_content = text if text else await extract_text(file) text_content = text if text else await extract_text(file)
summarized_text = await summarize_text(text_content, instruction) summarized_text = await summarize_text(text_content, instruction)
return summarized_text return summarized_text
@ -526,7 +526,7 @@ async def summarize_post(file: Optional[UploadFile] = File(None), text: Optional
@llm.post("/speaksummary") @llm.post("/speaksummary")
async def summarize_tts_endpoint( async def summarize_tts_endpoint(
bg_tasks: BackgroundTasks, bg_tasks: BackgroundTasks,
instruction: str = Form(SUMMARY_INSTRUCT), instruction: str = Form(Llm.summary.instruct),
file: Optional[UploadFile] = File(None), file: Optional[UploadFile] = File(None),
text: Optional[str] = Form(None), text: Optional[str] = Form(None),
voice: Optional[str] = Form(None), voice: Optional[str] = Form(None),
@ -572,7 +572,7 @@ async def summarize_tts_endpoint(
async def summarize_tts( async def summarize_tts(
text: str, text: str,
instruction: str = SUMMARY_INSTRUCT, instruction: str = Llm.summary.instruct,
voice: Optional[str] = None, voice: Optional[str] = None,
speed: float = 1.1, speed: float = 1.1,
podcast: bool = False, podcast: bool = False,
@ -605,9 +605,9 @@ def split_text_into_chunks(text: str) -> List[str]:
sentences = re.split(r'(?<=[.!?])\s+', text) sentences = re.split(r'(?<=[.!?])\s+', text)
words = text.split() words = text.split()
total_words = len(words) total_words = len(words)
l.debug(f"Total words: {total_words}. SUMMARY_CHUNK_SIZE: {SUMMARY_CHUNK_SIZE}. SUMMARY_TPW: {SUMMARY_TPW}.") l.debug(f"Total words: {total_words}. Llm.summary.chunk_size: {Llm.summary.chunk_size}. Llm.summary.length_ratio: {Llm.summary.length_ratio}.")
max_words_per_chunk = int(SUMMARY_CHUNK_SIZE / SUMMARY_TPW) max_words_per_chunk = int(Llm.summary.chunk_size / Llm.summary.length_ratio)
l.debug(f"Maximum words per chunk: {max_words_per_chunk}") l.debug(f"Maximum words per chunk: {max_words_per_chunk}")
chunks = [] chunks = []
@ -633,8 +633,8 @@ def split_text_into_chunks(text: str) -> List[str]:
def calculate_max_tokens(text: str) -> int: def calculate_max_tokens(text: str) -> int:
tokens_count = max(1, int(len(text.split()) * SUMMARY_TPW)) # Ensure at least 1 tokens_count = max(1, int(len(text.split()) * Llm.summary.length_ratio)) # Ensure at least 1
return min(tokens_count // 4, SUMMARY_CHUNK_SIZE) return min(tokens_count // 4, Llm.summary.chunk_size)
@ -694,7 +694,7 @@ async def extract_text(file: Union[UploadFile, bytes, bytearray, str, Path], bg_
raise ValueError(f"Error extracting text: {str(e)}") raise ValueError(f"Error extracting text: {str(e)}")
async def summarize_text(text: str, instruction: str = SUMMARY_INSTRUCT, length_override: int = None, length_quotient: float = SUMMARY_LENGTH_RATIO, LLM: Ollama = None): async def summarize_text(text: str, instruction: str = Llm.summary.instruct, length_override: int = None, length_quotient: float = Llm.summary.length_ratio, LLM: Ollama = None):
LLM = LLM if LLM else Ollama() LLM = LLM if LLM else Ollama()
chunked_text = split_text_into_chunks(text) chunked_text = split_text_into_chunks(text)
@ -703,12 +703,12 @@ async def summarize_text(text: str, instruction: str = SUMMARY_INSTRUCT, length_
total_words_count = sum(len(chunk.split()) for chunk in chunked_text) total_words_count = sum(len(chunk.split()) for chunk in chunked_text)
l.debug(f"Total words count: {total_words_count}") l.debug(f"Total words count: {total_words_count}")
total_tokens_count = max(1, int(total_words_count * SUMMARY_TPW)) total_tokens_count = max(1, int(total_words_count * Llm.summary.length_ratio))
l.debug(f"Total tokens count: {total_tokens_count}") l.debug(f"Total tokens count: {total_tokens_count}")
total_summary_length = length_override if length_override else total_tokens_count // length_quotient total_summary_length = length_override if length_override else total_tokens_count // length_quotient
l.debug(f"Total summary length: {total_summary_length}") l.debug(f"Total summary length: {total_summary_length}")
corrected_total_summary_length = min(total_summary_length, SUMMARY_TOKEN_LIMIT) corrected_total_summary_length = min(total_summary_length, Llm.summary.max_tokens)
l.debug(f"Corrected total summary length: {corrected_total_summary_length}") l.debug(f"Corrected total summary length: {corrected_total_summary_length}")
summaries = await asyncio.gather(*[ summaries = await asyncio.gather(*[
@ -738,11 +738,11 @@ async def process_chunk(instruction: str, text: str, part: int, total_parts: int
LLM = LLM if LLM else Ollama() LLM = LLM if LLM else Ollama()
words_count = len(text.split()) words_count = len(text.split())
tokens_count = max(1, int(words_count * SUMMARY_TPW)) tokens_count = max(1, int(words_count * Llm.summary.length_ratio))
summary_length_ratio = length_ratio if length_ratio else SUMMARY_LENGTH_RATIO Llm.summary.length_ratio = length_ratio if length_ratio else Llm.summary.length_ratio
max_tokens = min(tokens_count // summary_length_ratio, SUMMARY_CHUNK_SIZE) max_tokens = min(tokens_count // Llm.summary.length_ratio, Llm.summary.chunk_size)
max_tokens = max(max_tokens, SUMMARY_MIN_LENGTH) max_tokens = max(max_tokens, Llm.summary.min_length)
l.debug(f"Processing part {part} of {total_parts}: Words: {words_count}, Estimated tokens: {tokens_count}, Max output tokens: {max_tokens}") l.debug(f"Processing part {part} of {total_parts}: Words: {words_count}, Estimated tokens: {tokens_count}, Max output tokens: {max_tokens}")
@ -753,7 +753,7 @@ async def process_chunk(instruction: str, text: str, part: int, total_parts: int
l.info(f"Starting LLM.generate for part {part} of {total_parts}") l.info(f"Starting LLM.generate for part {part} of {total_parts}")
response = await LLM.generate( response = await LLM.generate(
model=SUMMARY_MODEL, model=Llm.summary.model,
prompt=prompt, prompt=prompt,
stream=False, stream=False,
options={'num_predict': max_tokens, 'temperature': 0.5} options={'num_predict': max_tokens, 'temperature': 0.5}