Merge branch 'main' into dependabot/pip/orjson-3.9.15

.pre-commit-config.yaml

@@ -7,7 +7,7 @@ repos:
       - id: trailing-whitespace
 
   - repo: https://github.com/psf/black
-    rev: 23.3.0
+    rev: 24.3.0
     hooks:
       - id: black-jupyter
       - id: black

azure/production/__main__.py

@@ -132,16 +132,6 @@
     resource_group_name=resource_group.name,
     restart_policy=containerinstance.ContainerGroupRestartPolicy.ALWAYS,
     sku=containerinstance.ContainerGroupSku.STANDARD,
-    volumes=[
-        containerinstance.VolumeArgs(
-            azure_file=containerinstance.AzureFileVolumeArgs(
-                share_name=file_share.name,
-                storage_account_key=storage_account_key,
-                storage_account_name=storage_account.name,
-            ),
-            name="llama-data",
-        ),
-    ],
 )
 
 # Define the container group for the data creation

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "reginald"
-version = "0.1.0"
+version= " 0.1.0"
 description = "A Slack bot for REG Hack Week 2023"
 authors = ["Evelina Gabasova <[email protected]>",
            "James Robinson <[email protected]>",
@@ -17,36 +17,44 @@ readme = "README.md"
 
 [tool.poetry.dependencies]
 python = ">=3.11,<3.12"
-accelerate = "^0.25.0"
+accelerate = "^0.28.0"
 bitsandbytes = { version="^0.41.1", optional=true }
 datasets = { version="^2.16.1", optional=true }
 faiss-cpu = { version="^1.7.4", optional=true }
-fastapi = { version="^0.103.1", optional=true }
-gitpython = "^3.1.41"
-gradio = { version = "^4.12.0", optional=true }
-httpx = "^0.26.0"
+fastapi = { version="^0.110.1", optional=true }
+gitpython = "^3.1.43"
+gradio = { version="^4.25.0", optional=true }
+httpx = "^0.27.0"
 ipykernel = { version="^6.23.2", optional=true }
-langchain = "^0.0.354"
-llama-cpp-python = "^0.2.27"
-llama-index = "^0.9.29"
-llama-hub = "^0.0.74"
-nbconvert = { version = "^7.8.0", optional = true }
-nest_asyncio = "^1.5.8"
-openai = "^1.6.1"
-pandas = "^2.0.2"
-pulumi = { version = "^3.100.0", optional=true }
-pulumi-azure-native = { version = "^2.24.0", optional=true }
-pydantic = { version = "^2.4.1", optional=true }
-requests = { version = "^2.31.0", optional=true }
-safetensors = "^0.3.3"
+langchain = "^0.1.14"
+llama-cpp-python = "^0.2.58"
+llama-index = "^0.10.26"
+nbconvert = { version="^7.8.0", optional=true }
+nest_asyncio = "^1.6.0"
+openai = "^1.16.0"
+pandas = "^2.2.1"
+pulumi = { version="^3.100.0", optional=true }
+pulumi-azure-native = { version="^2.24.0", optional=true }
+pydantic = { version="^2.4.1", optional=true }
+requests = { version="^2.31.0", optional=true }
+safetensors = "^0.4.2"
 slack-sdk = "^3.26.1"
-sentence-transformers = "^2.2.2"
-torch = "^2.1.2"
-transformers = "^4.36.2"
+sentence-transformers = "^2.6.1"
+torch = "^2.2.2"
+transformers = "^4.39.2"
 uvicorn = { version="^0.23.2", optional=true }
+llama-index-readers-github = "^0.1.7"
+llama-index-llms-azure-openai = "^0.1.5"
+llama-index-llms-openai = "^0.1.14"
+llama-index-llms-huggingface = "^0.1.4"
+llama-index-llms-llama-cpp = "^0.1.3"
+llama-index-readers-file = "^0.1.13"
+llama-index-embeddings-langchain = "^0.1.2"
+langchain-community = "^0.0.31"
+tiktoken = "^0.6.0"
 
 [tool.poetry.group.dev.dependencies]
-black = "^23.3.0"
+black = "^24.3.0"
 isort = "^5.12.0"
 pre-commit = "^3.3.2"
 

reginald/models/create_index.py

@@ -2,23 +2,22 @@
 import pathlib
 from typing import Any
 
-from llama_index.llms import (
+from llama_index.core.base.llms.types import (
     CompletionResponse,
     CompletionResponseGen,
-    CustomLLM,
     LLMMetadata,
 )
-from llama_index.llms.base import llm_completion_callback
-from llama_index.llms.custom import CustomLLM
+from llama_index.core.llms.callbacks import llm_completion_callback
+from llama_index.core.llms.custom import CustomLLM
 
-from reginald.models.models.llama_index import DataIndexCreator, setup_service_context
+from reginald.models.models.llama_index import DataIndexCreator, setup_settings
 from reginald.models.setup_llm import DEFAULT_ARGS
 from reginald.parser_utils import Parser, get_args
 
 
 class DummyLLM(CustomLLM):
     """
-    Dummy LLM for passing into the ServiceContext below to create the index.
+    Dummy LLM for passing into the Settings below to create the index.
     The minimum required attributes are set here, but this LLM is not used anywhere else.
     """
 
@@ -65,8 +64,8 @@ def main():
     args = get_args(parser)
 
     # pass args to create data index
-    logging.info("Setting up service context...")
-    service_context = setup_service_context(
+    logging.info("Setting up settings...")
+    settings = setup_settings(
         llm=DummyLLM(),
         max_input_size=args.max_input_size or DEFAULT_ARGS["max_input_size"],
         num_output=args.num_output or DEFAULT_ARGS["num_output"],
@@ -81,7 +80,7 @@ def main():
     data_creator = DataIndexCreator(
         data_dir=pathlib.Path(args.data_dir or DEFAULT_ARGS["data_dir"]).resolve(),
         which_index=args.which_index or DEFAULT_ARGS["which_index"],
-        service_context=service_context,
+        settings=settings,
     )
     data_creator.create_index()
     data_creator.save_index()