Merge pull request #1 from curieo-org/dockerization

Dockerize search app

.github/workflows/tests.yaml

@@ -2,15 +2,10 @@ name: Tests
 
 on:
   push:
-  pull_request:
 
 jobs:
   tests:
     runs-on: ubuntu-latest
-    strategy:
-      max-parallel: 4
-      matrix:
-        python-version: ["3.10", "3.11"]
 
     services:
       postgres:
@@ -24,19 +19,15 @@ jobs:
         options: --health-cmd pg_isready --health-interval 5s --health-timeout 5s --health-retries 5
 
     steps:
-    - uses: actions/checkout@v2
-
-    - name: Setup Python ${{ matrix.python-version }}
-      uses: actions/setup-python@v2
+    - uses: actions/checkout@v4
+    - name: Install poetry
+      run: pipx install poetry
+    - uses: actions/setup-python@v5
       with:
-        python-version: ${{ matrix.python-version }}
-        cache: 'pip'
-        cache-dependency-path: '**/requirements/development.txt'
+        cache: 'poetry'
 
     - name: Install dependencies
-      run: |
-        python -m pip install --upgrade pip
-        pip install -r app/requirements/development.txt
+      run: poetry install
 
     - name: Run tests
       env:
@@ -45,4 +36,4 @@ jobs:
         DB_USER: postgres
         DB_PASS: postgres
         DB_BASE: db_test
-      run: cd app && pytest .
+      run: cd app && poetry run pytest

.gitignore

@@ -158,4 +158,4 @@ cython_debug/
 #  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
 #  and can be added to the global gitignore or merged into this file.  For a more nuclear
 #  option (not recommended) you can uncomment the following to ignore the entire idea folder.
-#.idea/
+.idea/

.python-version

@@ -0,0 +1 @@
+3.11

Dockerfile

@@ -0,0 +1,20 @@
+# pull official base image
+FROM python:3.11-slim
+
+# install system dependencies
+RUN apt-get update && apt-get -y install libpq-dev gcc
+
+
+# install python dependencies
+COPY ./pyproject.toml /
+COPY ./poetry.lock /
+RUN pip install poetry uvicorn
+RUN poetry config virtualenvs.create false
+RUN poetry install --compile
+
+COPY ./app /app
+COPY .env /.env
+
+EXPOSE 8000
+
+CMD ["uvicorn", "app.main:app", "--host=0.0.0.0", "--port", "80"]

README.md

@@ -82,15 +82,16 @@ The project might have multiple branches: `master`, `development`, etc. which ca
 
 ```
 
-### Tools Required
-
--   A text editor or an IDE (VsCode is Preferred)
-
 ### Installation
 
 All installation steps go here.
 
--   Install `poetry` for the dependencies. - `poetry install`
+- Make sure `pyenv` is installed.
+- Create a new virtual environment `python -m venv .venv`
+- Activate the env `source .venv/bin/activate`
+- Install poetry and uvicorn `pip install poetry uvicorn`
+- Install dependencies `poetry install`
+- Now you can run the project with `uvicorn app.main:app --reload`
 
 -   Set up Embedding Server from [TEI](https://github.com/huggingface/text-embeddings-inference/tree/main)
 

app/rag/retrieval/clinical_trials/clinical_trial_sql_query_engine.py

@@ -19,15 +19,12 @@
 from typing import List
 from sqlalchemy import create_engine
 from pyvis.network import Network
-import phoenix as px
 
 from app.config import OPENAPI_KEY, CLINICAL_TRIALS_TABLE_INFO_DIR, POSTGRES_ENGINE, EMBEDDING_MODEL_API, EMBEDDING_MODEL_NAME
 from app.services.search_utility import setup_logger
 
 logger = setup_logger('ClinicalTrialText2SQLEngine')
 
-#px.launch_app(port=6060)
-
 class TableInfo(BaseModel):
     """
     Information regarding a structured table.
@@ -47,7 +44,7 @@ class ClinicalTrialText2SQLEngine:
     """
     def __init__(self, config):
         self.config = config
-        
+
         self.llm = OpenAI(model="gpt-3.5-turbo", api_key=str(OPENAPI_KEY))
         self.engine = create_engine(str(POSTGRES_ENGINE))
 
@@ -56,8 +53,8 @@ def __init__(self, config):
         self.sql_database = SQLDatabase(self.engine)
         self.table_node_mapping = SQLTableNodeMapping(self.sql_database)
         self.sql_retriever = SQLRetriever(self.sql_database)
-        
-        self.table_schema_objs = [SQLTableSchema(table_name=t.table_name, context_str=t.table_summary) for t in self.get_all_table_info()] 
+
+        self.table_schema_objs = [SQLTableSchema(table_name=t.table_name, context_str=t.table_summary) for t in self.get_all_table_info()]
         self.embed_model = TextEmbeddingsInference(base_url=EMBEDDING_MODEL_API, model_name=EMBEDDING_MODEL_NAME)
 
         self.obj_index = ObjectIndex.from_objects(objects=self.table_schema_objs, object_mapping=self.table_node_mapping, index_cls=VectorStoreIndex, embed_model=self.embed_model)
@@ -85,7 +82,7 @@ def _get_table_info_with_index(self, idx: int) -> str:
             raise ValueError(
                 f"More than one file matching index: {list(results_gen)}"
             )
-        
+
     def get_all_table_info(self):
         file_counts = len(os.listdir(CLINICAL_TRIALS_TABLE_INFO_DIR))
         table_infos = []
@@ -122,7 +119,7 @@ def parse_response_to_sql(self, response: ChatResponse) -> str:
             response = response[:sql_result_start]
         logger.debug(f"ClinicalTrialText2SQLEngine.parse_response_to_sql sql: {response}")
         return response.strip().strip("```").strip()
-    
+
     def get_response_synthesis_prompt(self, query_str, sql_query, context_str) -> PromptTemplate:
         response_synthesis_prompt_str = (
             "Given an input question, synthesize a response from the query results.\n"
@@ -179,7 +176,7 @@ def call_text2sql(self, search_text:str):
         except Exception as ex:
             logger.exception("ClinicalTrialText2SQLEngine.call_text2sql Exception -", exc_info = ex, stack_info=True)
             raise ex
-        
+
         return {
             "result" : str(response)
         }

app/tests/test_stub.py

@@ -0,0 +1,2 @@
+def test_placeholder():
+    pass

docker-compose.yml

@@ -1,14 +1,10 @@
 version: '3.8'
 
 services:
-
   web:
     build:
       context: .
-      dockerfile: docker/Dockerfile.local
-    command: ./scripts/docker-entrypoint.sh
-    volumes:
-    - ./app:/usr/src/app
+      dockerfile: Dockerfile
     ports:
     - 8000:8000
     depends_on: