Update requirements fix warnings

rag_experiment_accelerator/config/config.py

@@ -124,7 +124,9 @@ def _initialize(self, config_dir: str, data_dir: str, filename: str) -> None:
         self.AzureSearchCredentials = AzureSearchCredentials.from_env()
         self.AzureMLCredentials = AzureMLCredentials.from_env()
         self.AzureSkillsCredentials = AzureSkillsCredentials.from_env()
-        self.AzureDocumentIntelligenceCredentials = AzureDocumentIntelligenceCredentials.from_env()
+        self.AzureDocumentIntelligenceCredentials = (
+            AzureDocumentIntelligenceCredentials.from_env()
+        )
 
         self.embedding_models: list[EmbeddingModel] = []
         embedding_model_config = data.get("embedding_models", [])
@@ -145,11 +147,11 @@ def _initialize(self, config_dir: str, data_dir: str, filename: str) -> None:
 
             self.MAIN_PROMPT_INSTRUCTION = data["main_prompt_instruction"]
             if self.MAIN_PROMPT_INSTRUCTION is None:
-                logger.warn(
+                logger.warning(
                     "prompt_config.json found but main_prompt_instruction is"
                     " not set. Using default prompts"
                 )
                 self.MAIN_PROMPT_INSTRUCTION = main_prompt_instruction
         except OSError:
-            logger.warn("prompt_config.json not found. Using default prompts")
+            logger.warning("prompt_config.json not found. Using default prompts")
             self.MAIN_PROMPT_INSTRUCTION = main_prompt_instruction

rag_experiment_accelerator/doc_loader/docxLoader.py

@@ -1,10 +1,12 @@
-from langchain.document_loaders import Docx2txtLoader
+from langchain_community.document_loaders import Docx2txtLoader
 
 from rag_experiment_accelerator.doc_loader.structuredLoader import (
     load_structured_files,
 )
 from rag_experiment_accelerator.utils.logging import get_logger
-from rag_experiment_accelerator.config.credentials import AzureDocumentIntelligenceCredentials
+from rag_experiment_accelerator.config.credentials import (
+    AzureDocumentIntelligenceCredentials,
+)
 
 logger = get_logger(__name__)
 

rag_experiment_accelerator/doc_loader/htmlLoader.py

@@ -1,10 +1,12 @@
-from langchain.document_loaders import BSHTMLLoader
+from langchain_community.document_loaders import BSHTMLLoader
 
 from rag_experiment_accelerator.doc_loader.structuredLoader import (
     load_structured_files,
 )
 from rag_experiment_accelerator.utils.logging import get_logger
-from rag_experiment_accelerator.config.credentials import AzureDocumentIntelligenceCredentials
+from rag_experiment_accelerator.config.credentials import (
+    AzureDocumentIntelligenceCredentials,
+)
 
 logger = get_logger(__name__)
 

rag_experiment_accelerator/doc_loader/markdownLoader.py

@@ -1,10 +1,12 @@
-from langchain.document_loaders import UnstructuredMarkdownLoader
+from langchain_community.document_loaders import UnstructuredMarkdownLoader
 
 from rag_experiment_accelerator.doc_loader.structuredLoader import (
     load_structured_files,
 )
 from rag_experiment_accelerator.utils.logging import get_logger
-from rag_experiment_accelerator.config.credentials import AzureDocumentIntelligenceCredentials
+from rag_experiment_accelerator.config.credentials import (
+    AzureDocumentIntelligenceCredentials,
+)
 
 logger = get_logger(__name__)
 

rag_experiment_accelerator/doc_loader/pdfLoader.py

@@ -1,9 +1,13 @@
-from langchain.document_loaders import PyPDFDirectoryLoader
+from langchain_community.document_loaders import PyPDFDirectoryLoader
 from langchain.text_splitter import RecursiveCharacterTextSplitter
 
 from rag_experiment_accelerator.utils.logging import get_logger
-from rag_experiment_accelerator.doc_loader.documentIntelligenceLoader import azure_document_intelligence_directory_loader
-from rag_experiment_accelerator.config.credentials import AzureDocumentIntelligenceCredentials
+from rag_experiment_accelerator.doc_loader.documentIntelligenceLoader import (
+    azure_document_intelligence_directory_loader,
+)
+from rag_experiment_accelerator.config.credentials import (
+    AzureDocumentIntelligenceCredentials,
+)
 import uuid
 import re
 
@@ -29,9 +33,9 @@ def preprocess_pdf_content(content: str):
         # Output: "hello world openai"
     """
 
-    content = re.sub(r'\n{2,}', '\n', content)
-    content = re.sub(r'\n{1,}', '', content)
-    content = re.sub(r'\\u[0-9a-fA-F]{4}', '', content)
+    content = re.sub(r"\n{2,}", "\n", content)
+    content = re.sub(r"\n{1,}", "", content)
+    content = re.sub(r"\\u[0-9a-fA-F]{4}", "", content)
     content = content.lower()
 
     return content
@@ -63,7 +67,12 @@ def load_pdf_files(
     documents = []
     for pattern in glob_patterns:
         if chunking_strategy == "azure-document-intelligence":
-            documents += azure_document_intelligence_directory_loader(pattern, folder_path, AzureDocumentIntelligenceCredentials.AZURE_DOCUMENT_INTELLIGENCE_ENDPOINT, AzureDocumentIntelligenceCredentials.AZURE_DOCUMENT_INTELLIGENCE_ADMIN_KEY)
+            documents += azure_document_intelligence_directory_loader(
+                pattern,
+                folder_path,
+                AzureDocumentIntelligenceCredentials.AZURE_DOCUMENT_INTELLIGENCE_ENDPOINT,
+                AzureDocumentIntelligenceCredentials.AZURE_DOCUMENT_INTELLIGENCE_ADMIN_KEY,
+            )
         else:
             # using langchain
             loader = PyPDFDirectoryLoader(
@@ -86,7 +95,9 @@ def load_pdf_files(
     docs = text_splitter.split_documents(documents)
     docsList = []
     for doc in docs:
-        docsList.append(dict({str(uuid.uuid4()): preprocess_pdf_content(doc.page_content)}))
+        docsList.append(
+            dict({str(uuid.uuid4()): preprocess_pdf_content(doc.page_content)})
+        )
 
     logger.info(f"Split {len(documents)} PDF pages into {len(docs)} chunks")
 

rag_experiment_accelerator/doc_loader/textLoader.py

@@ -1,10 +1,12 @@
-from langchain.document_loaders import TextLoader
+from langchain_community.document_loaders import TextLoader
 
 from rag_experiment_accelerator.doc_loader.structuredLoader import (
     load_structured_files,
 )
 from rag_experiment_accelerator.utils.logging import get_logger
-from rag_experiment_accelerator.config.credentials import AzureDocumentIntelligenceCredentials
+from rag_experiment_accelerator.config.credentials import (
+    AzureDocumentIntelligenceCredentials,
+)
 
 logger = get_logger(__name__)
 

rag_experiment_accelerator/init_Index/create_index.py

@@ -35,7 +35,6 @@ def create_acs_index(
     analyzers,
 ):
     try:
-
         credential = AzureKeyCredential(key)
 
         # Apply checks on analyzer settings. Search analyzer and index analyzer must be set together
@@ -44,15 +43,20 @@ def create_acs_index(
 
         if bool(index_analyzer_name) != bool(search_analyzer_name):
             raise ValueError(
-                "Both 'index_analyzer_name' and 'search_analyzer_name' must be set together")
+                "Both 'index_analyzer_name' and 'search_analyzer_name' must be set together"
+            )
 
         index_analyzer = index_analyzer_name if index_analyzer_name else ""
         search_analyzer = search_analyzer_name if search_analyzer_name else ""
 
         # Analyzer can only be used if neither search analyzer or index analyzer are set
-        if analyzers.get("analyzer_name") and (analyzers.get("search_analyzer_name") or analyzers.get("index_analyzer_name")):
+        if analyzers.get("analyzer_name") and (
+            analyzers.get("search_analyzer_name")
+            or analyzers.get("index_analyzer_name")
+        ):
             raise ValueError(
-                "analyzer_name should be empty if either search_analyzer_name or index_analyzer_name is not empty")
+                "analyzer_name should be empty if either search_analyzer_name or index_analyzer_name is not empty"
+            )
         analyzer = analyzers.get("analyzer_name") or ""
 
         # Create a search index
@@ -100,24 +104,21 @@ def create_acs_index(
             ),
             SearchField(
                 name="contentVector",
-                type=SearchFieldDataType.Collection(
-                    SearchFieldDataType.Single),
+                type=SearchFieldDataType.Collection(SearchFieldDataType.Single),
                 searchable=True,
                 vector_search_dimensions=int(dimension),
                 vector_search_profile="my-vector-search-profile",
             ),
             SearchField(
                 name="contentTitle",
-                type=SearchFieldDataType.Collection(
-                    SearchFieldDataType.Single),
+                type=SearchFieldDataType.Collection(SearchFieldDataType.Single),
                 searchable=True,
                 vector_search_dimensions=int(dimension),
                 vector_search_profile="my-vector-search-profile",
             ),
             SearchField(
                 name="contentSummary",
-                type=SearchFieldDataType.Collection(
-                    SearchFieldDataType.Single),
+                type=SearchFieldDataType.Collection(SearchFieldDataType.Single),
                 searchable=True,
                 vector_search_dimensions=int(dimension),
                 vector_search_profile="my-vector-search-profile",
@@ -154,8 +155,7 @@ def create_acs_index(
         semantic_config = SemanticConfiguration(
             name="my-semantic-config",
             prioritized_fields=PrioritizedFields(
-                prioritized_content_fields=[
-                    SemanticField(field_name="content")]
+                prioritized_content_fields=[SemanticField(field_name="content")]
             ),
         )
 
@@ -190,8 +190,7 @@ def create_acs_index(
                 for char_filter in analyzers["char_filters"]
             ]
 
-        cors_options = CorsOptions(
-            allowed_origins=["*"], max_age_in_seconds=60)
+        cors_options = CorsOptions(allowed_origins=["*"], max_age_in_seconds=60)
         scoring_profiles = []
 
         # Create the search index with the semantic, tokenizer, and filter settings