sij/khoj
1
0
Fork 0
mirror of https://github.com/khoj-ai/khoj.git synced 2024-11-23 15:38:55 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions 5

Migrate to OpenAI Python library >= 1.0

Browse source
This commit is contained in:
Debanjum Singh Solanky 2023-12-03 18:16:00 -05:00
parent a57d529f39
commit 7009793170
4 changed files with 16 additions and 15 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
pyproject.toml
View file

@ -42,7 +42,7 @@ dependencies = [
"fastapi >= 0.104.1", "fastapi >= 0.104.1",
"python-multipart >= 0.0.5", "python-multipart >= 0.0.5",
"jinja2 == 3.1.2", "jinja2 == 3.1.2",
"openai >= 0.27.0, < 1.0.0", "openai >= 1.0.0",
"tiktoken >= 0.3.2", "tiktoken >= 0.3.2",
"tenacity >= 8.2.2", "tenacity >= 8.2.2",
"pillow ~= 9.5.0", "pillow ~= 9.5.0",

20
src/khoj/processor/conversation/openai/utils.py
View file

@ -36,11 +36,11 @@ class StreamingChatCallbackHandler(StreamingStdOutCallbackHandler):
@retry( @retry(
retry=( retry=(
retry_if_exception_type(openai.error.Timeout) retry_if_exception_type(openai._exceptions.APITimeoutError)
| retry_if_exception_type(openai.error.APIError) | retry_if_exception_type(openai._exceptions.APIError)
| retry_if_exception_type(openai.error.APIConnectionError) | retry_if_exception_type(openai._exceptions.APIConnectionError)
| retry_if_exception_type(openai.error.RateLimitError) | retry_if_exception_type(openai._exceptions.RateLimitError)
| retry_if_exception_type(openai.error.ServiceUnavailableError) | retry_if_exception_type(openai._exceptions.APIStatusError)
), ),
wait=wait_random_exponential(min=1, max=10), wait=wait_random_exponential(min=1, max=10),
stop=stop_after_attempt(3), stop=stop_after_attempt(3),
@ -57,11 +57,11 @@ def completion_with_backoff(**kwargs):
@retry( @retry(
retry=( retry=(
retry_if_exception_type(openai.error.Timeout) retry_if_exception_type(openai._exceptions.APITimeoutError)
| retry_if_exception_type(openai.error.APIError) | retry_if_exception_type(openai._exceptions.APIError)
| retry_if_exception_type(openai.error.APIConnectionError) | retry_if_exception_type(openai._exceptions.APIConnectionError)
| retry_if_exception_type(openai.error.RateLimitError) | retry_if_exception_type(openai._exceptions.RateLimitError)
| retry_if_exception_type(openai.error.ServiceUnavailableError) | retry_if_exception_type(openai._exceptions.APIStatusError)
), ),
wait=wait_exponential(multiplier=1, min=4, max=10), wait=wait_exponential(multiplier=1, min=4, max=10),
stop=stop_after_attempt(3), stop=stop_after_attempt(3),

5
src/khoj/processor/conversation/openai/whisper.py
View file

@ -3,7 +3,7 @@ from io import BufferedReader
# External Packages # External Packages
from asgiref.sync import sync_to_async from asgiref.sync import sync_to_async
import openai from openai import OpenAI
async def transcribe_audio(audio_file: BufferedReader, model, api_key) -> str: async def transcribe_audio(audio_file: BufferedReader, model, api_key) -> str:
@ -11,5 +11,6 @@ async def transcribe_audio(audio_file: BufferedReader, model, api_key) -> str:
Transcribe audio file using Whisper model via OpenAI's API Transcribe audio file using Whisper model via OpenAI's API
""" """
# Send the audio data to the Whisper API # Send the audio data to the Whisper API
response = await sync_to_async(openai.Audio.translate)(model=model, file=audio_file, api_key=api_key) client = OpenAI(api_key=api_key)
response = await sync_to_async(client.audio.translations.create)(model=model, file=audio_file)
return response["text"] return response["text"]

4
src/khoj/utils/models.py
View file

@ -32,7 +32,7 @@ class OpenAI(BaseEncoder):
raise Exception( raise Exception(
f"Set OpenAI API key under processor-config > conversation > openai-api-key in config file: {state.config_file}" f"Set OpenAI API key under processor-config > conversation > openai-api-key in config file: {state.config_file}"
) )
openai.api_key = state.processor_config.conversation.openai_model.api_key self.openai_client = openai.OpenAI(api_key=state.processor_config.conversation.openai_model.api_key)
self.embedding_dimensions = None self.embedding_dimensions = None
def encode(self, entries, device=None, **kwargs): def encode(self, entries, device=None, **kwargs):
@ -43,7 +43,7 @@ class OpenAI(BaseEncoder):
processed_entry = entries[index].replace("\n", " ") processed_entry = entries[index].replace("\n", " ")
try: try:
response = openai.Embedding.create(input=processed_entry, model=self.model_name) response = self.openai_client.embeddings.create(input=processed_entry, model=self.model_name)
embedding_tensors += [torch.tensor(response.data[0].embedding, device=device)] embedding_tensors += [torch.tensor(response.data[0].embedding, device=device)]
# Use current models embedding dimension, once available # Use current models embedding dimension, once available
# Else default to embedding dimensions of the text-embedding-ada-002 model # Else default to embedding dimensions of the text-embedding-ada-002 model