Linting

README.md

@@ -133,6 +133,14 @@ Pre commit hooks run after commit to fix up formatting and other issues. Install
 pre-commit install
 ```
 
+You can then run:
+
+```sh
+pre-commit run --all-files
+```
+
+and commit any changes made to files in the repo.
+
 ## 6. Add secrets into .env
 
 - Run `cp .env.template .env` and update the secrets.

app/home.py

@@ -8,9 +8,7 @@
 
 from config.logging import setup_logging
 from config.settings import ENV
-from hackathon.streamlit.utils import (
-    check_password,
-)
+from hackathon.streamlit.utils import check_password
 
 get_logger = setup_logging()
 logger = get_logger(__name__)
@@ -155,4 +153,3 @@ def image_to_base64(image):
 """
 )
 # * [Initialise the data](/initialise_the_data) will load in the previous vacancies from s3.
-

config/settings.py

@@ -9,9 +9,7 @@
 PROJECT_PATH = os.environ.get("PROJECT_PATH")
 OPENSEARCH_URL = os.environ.get("OPENSEARCH_URL")
 OPENSEARCH_BATCH_SIZE = int(os.environ.get("OPENSEARCH_BATCH_SIZE", 500))
-OPENSEARCH_ENDPOINT_NAME = os.environ.get(
-    "OPENSEARCH_ENDPOINT_NAME", "vstore"
-)
+OPENSEARCH_ENDPOINT_NAME = os.environ.get("OPENSEARCH_ENDPOINT_NAME", "vstore")
 OPENSEARCH_INDEX_NAME = os.environ.get("OPENSEARCH_INDEX_NAME", "vacancies")
 OPENSEARCH_SKILLS_INDEX_NAME = os.environ.get(
     "OPENSEARCH_SKILLS_INDEX_NAME", "unique_skills"

hackathon/llm/chain_config.py

@@ -5,10 +5,7 @@
 from langchain.schema.output_parser import StrOutputParser
 from langchain_core.output_parsers.transform import BaseTransformOutputParser
 
-from hackathon.llm.prompts.core import (
-
-    PromptTemplate,
-)
+from hackathon.llm.prompts.core import PromptTemplate
 
 
 @dataclass

hackathon/llm/llm_handler.py

@@ -16,7 +16,10 @@ class LLMRunner:
     """
 
     def __init__(
-        self, llm: LLM, vectorstore: VectorStore, chain_configs: List[ChainConfig]
+        self,
+        llm: LLM,
+        vectorstore: VectorStore,
+        chain_configs: List[ChainConfig],
     ) -> None:
         """
         Sets up runner with required components of an llm, vectorstore and a list of llm chain configs.

hackathon/llm/prompts/core.py

@@ -39,4 +39,3 @@
     """
 """
 )
-

hackathon/streamlit/utils.py

@@ -33,9 +33,10 @@
     S3_LOADER_FILE_NAME,
     VECTOR_STORE_CONFIG,
 )
-from hackathon.llm.chain_config import (
 
-)
+# from hackathon.llm.chain_config import (
+# # Empty for now
+# )
 from hackathon.llm.llm import LLama2, SagemakerHostedLLM
 from hackathon.llm.llm_handler import LLMRunner
 from hackathon.loader.chunker import TextChunker
@@ -128,20 +129,25 @@ def initialise_llm_runner():
 
     if VECTOR_STORE_CONFIG == "chroma":
         vector_store = ChromaStore(
-            embedding_function=st_embedder, collection_name=OPENSEARCH_INDEX_NAME
+            embedding_function=st_embedder,
+            collection_name=OPENSEARCH_INDEX_NAME,
         )
     else:
         if "skills_os_client" not in st.session_state:
             st.session_state["skills_os_client"] = OpensearchClient(
-                OPENSEARCH_SKILLS_INDEX_NAME, OPENSEARCH_ENDPOINT_NAME, AWS_REGION
+                OPENSEARCH_SKILLS_INDEX_NAME,
+                OPENSEARCH_ENDPOINT_NAME,
+                AWS_REGION,
             )
         if "vacancy_os_client" not in st.session_state:
             st.session_state["vacancy_os_client"] = OpensearchClient(
                 OPENSEARCH_INDEX_NAME, OPENSEARCH_ENDPOINT_NAME, AWS_REGION
             )
 
         vector_store = OpenSearchStore(
-            st_embedder, OPENSEARCH_INDEX_NAME, st.session_state["vacancy_os_client"]
+            st_embedder,
+            OPENSEARCH_INDEX_NAME,
+            st.session_state["vacancy_os_client"],
         )
 
     if LLM_MODEL == "local_llm":
@@ -156,11 +162,11 @@ def initialise_llm_runner():
     llm_runner = LLMRunner(
         llm=llm,
         vectorstore=vector_store,
-        chain_configs=[
-        ],
+        chain_configs=[],
     )
     st.session_state["runner"] = llm_runner
 
+
 def initialise_vector_store_loader():
     if LLM_MODEL == "local_llm":
         st_embedder = SentenceTransformerEmbeddings(
@@ -179,7 +185,9 @@ def initialise_vector_store_loader():
                 OPENSEARCH_INDEX_NAME, OPENSEARCH_ENDPOINT_NAME, AWS_REGION
             )
         vector_store = OpensearchClientStore(
-            st_embedder, OPENSEARCH_INDEX_NAME, st.session_state["vacancy_os_client"]
+            st_embedder,
+            OPENSEARCH_INDEX_NAME,
+            st.session_state["vacancy_os_client"],
         )
 
     if LOADER_CONFIG == "file_loader":
@@ -197,9 +205,9 @@ def initialise_vector_store_loader():
         chunker=TextChunker(chunk_size=1000, overlap=10),
     )
 
+
 def safe_literal_eval(x):
     try:
         return literal_eval(x)
     except (SyntaxError, ValueError):
         return None
-

hackathon/vectorstore/vectorstore.py

@@ -115,7 +115,12 @@ def retrieve_data_with_score(self, query):
         return self.vectorstore.similarity_search_with_score(query, k=10)
 
     def retrieve_data(
-        self, query: str, search_type: str, space_type: str, pre_filter, k: int = 5
+        self,
+        query: str,
+        search_type: str,
+        space_type: str,
+        pre_filter,
+        k: int = 5,
     ):
         """
         Similarity search which returns values with score attachmend with up to k results,