Upload 8 files

.dockerignore

@@ -0,0 +1,21 @@
+# Exclude files not needed for Docker build
+*__pycache__/
+*.pyc
+*.pyo
+
+# Exclude log files but keep the logs directory
+logs/*
+!logs/.gitkeep
+
+# Exclude notebook checkpoints
+*.ipynb_checkpoints/
+
+# Environments
+.env
+*.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/

Dockerfile-api

@@ -0,0 +1,23 @@
+FROM python:3.11.9-slim
+
+# Copy requirements file
+COPY requirements_docker_api.txt .
+
+# Update pip
+RUN pip --timeout=30000 install --no-cache-dir --upgrade pip
+
+# Install dependecies
+RUN pip --timeout=30000 install --no-cache-dir -r requirements_docker_api.txt
+
+# Copy API
+RUN mkdir /src/
+COPY ./src /src
+
+# Set workdir
+WORKDIR /
+
+# Expose app port
+EXPOSE 9000
+
+# Start application
+CMD ["uvicorn", "src.api.api:app", "--host", "0.0.0.0", "--port", "9000", "--reload"]

requirements_docker_api.txt

@@ -0,0 +1,85 @@
+aiohttp==3.9.5
+aiosignal==1.3.1
+annotated-types==0.7.0
+anyio==4.4.0
+asttokens==2.4.1
+attrs==23.2.0
+certifi==2024.7.4
+click==8.1.7
+colorama==0.4.6
+comm==0.2.2
+contourpy==1.2.1
+cycler==0.12.1
+debugpy==1.6.7
+decorator==5.1.1
+dnspython==2.6.1
+email_validator==2.2.0
+exceptiongroup==1.2.0
+executing==2.0.1
+fastapi==0.111.0
+fastapi-cache2==0.2.1
+fastapi-cli==0.0.4
+fonttools==4.53.1
+frozenlist==1.4.1
+h11==0.14.0
+httpcore==1.0.5
+httptools==0.6.1
+httpx==0.27.0
+idna==3.7
+imbalanced-learn==0.12.3
+imblearn==0.0
+importlib_metadata==8.0.0
+jedi==0.19.1
+Jinja2==3.1.4
+joblib==1.4.2
+mdurl==0.1.2
+multidict==6.0.5
+nest_asyncio==1.6.0
+numpy==2.0.0
+orjson==3.10.6
+packaging==24.1
+pandas==2.2.2
+parso==0.8.4
+pendulum==3.0.0
+pickleshare==0.7.5
+pillow==10.4.0
+pip==24.0
+platformdirs==4.2.2
+prompt_toolkit==3.0.47
+pure-eval==0.2.2
+pydantic==2.8.2
+pydantic_core==2.20.1
+Pygments==2.18.0
+pyparsing==3.1.2
+python-dateutil==2.9.0
+python-dotenv==1.0.1
+python-multipart==0.0.9
+pytz==2024.1
+PyYAML==6.0.1
+pyzmq==25.1.2
+redis==5.0.7
+rich==13.7.1
+scikit-learn==1.5.1
+scipy==1.14.0
+shellingham==1.5.4
+six==1.16.0
+sniffio==1.3.1
+stack-data==0.6.2
+starlette==0.37.2
+tenacity==8.5.0
+threadpoolctl==3.5.0
+time-machine==2.14.2
+tornado==6.4.1
+traitlets==5.14.3
+typer==0.12.3
+typing_extensions==4.12.2
+tzdata==2024.1
+ujson==5.10.0
+uvicorn==0.30.1
+watchfiles==0.22.0
+wcwidth==0.2.13
+websockets==12.0
+wheel==0.43.0
+xgboost==2.1.0
+yarl==1.9.4
+zipp==3.19.2

src/api/.env

@@ -0,0 +1,3 @@
+REDIS_URL=redis://redis-14458.c14.us-east-1-2.ec2.redns.redis-cloud.com:14458/
+REDIS_USERNAME=default
+REDIS_PASSWORD=lkiTwONUyGZdOlg216hwOAV3L82OJuRM

src/api/api.py

@@ -0,0 +1,161 @@
+import os
+from dotenv import load_dotenv
+
+from collections.abc import AsyncIterator
+from contextlib import asynccontextmanager
+
+from fastapi import FastAPI
+from fastapi_cache import FastAPICache
+from fastapi_cache.backends.redis import RedisBackend
+from fastapi_cache.coder import PickleCoder
+from fastapi_cache.decorator import cache
+
+from redis import asyncio as aioredis
+
+from pydantic import BaseModel
+from typing import Tuple, Dict, Union
+
+from imblearn.pipeline import Pipeline as imbPipeline
+from sklearn.preprocessing._label import LabelEncoder
+import joblib
+import pandas as pd
+from urllib.request import urlopen
+
+
+from src.api.config import ONE_DAY_SEC, ONE_WEEK_SEC, XGBOOST_URL, RANDOM_FOREST_URL, ENCODER_URL, ENV_PATH
+
+load_dotenv(ENV_PATH)
+
+
+@asynccontextmanager
+async def lifespan(_: FastAPI) -> AsyncIterator[None]:
+    url = os.getenv("REDIS_URL")
+    username = os.getenv("REDIS_USERNAME")
+    password = os.getenv("REDIS_PASSWORD")
+    redis = aioredis.from_url(url=url, username=username,
+                              password=password, encoding="utf8", decode_responses=True)
+    FastAPICache.init(RedisBackend(redis), prefix="fastapi-cache")
+    yield
+
+
+# FastAPI Object
+app = FastAPI(
+    title='Sepsis classification',
+    version='0.0.1',
+    description='Identify ICU patients at risk of developing sepsis',
+    lifespan=lifespan,
+)
+
+
+# API input features
+class SepsisFeatures(BaseModel):
+    PRG: float
+    PL: float
+    PR: float
+    SK: float
+    TS: float
+    M11: float
+    BD2: float
+    Age: float
+    Insurance: float
+
+
+class Url(BaseModel):
+    pipeline_url: str
+    encoder_url: str
+
+
+class ResultData(BaseModel):
+    prediction: str
+    probability: float
+
+
+class PredictionResponse(BaseModel):
+    execution_msg: str
+    execution_code: int
+    result: ResultData
+
+
+class ErrorResponse(BaseModel):
+    execution_msg: Union[str, None]
+    execution_code: Union[int, None]
+    result: Union[Dict[str, Union[str, int]], Union[Dict[str, None], None]]
+
+
+# Load the model pipelines and encoder
+# Cache for 1 day
+@cache(expire=ONE_DAY_SEC, namespace='pipeline_resource', coder=PickleCoder)
+async def load_pipeline(pipeline_url: Url, encoder_url: Url) -> Tuple[imbPipeline, LabelEncoder]:
+    pipeline, encoder = None, None
+    try:
+        pipeline: imbPipeline = joblib.load(urlopen(pipeline_url))
+        encoder: LabelEncoder = joblib.load(urlopen(encoder_url))
+    except Exception:
+        # Log exception
+        pass
+    finally:
+        return pipeline, encoder
+
+
+# Endpoints
+
+# Status endpoint: check if api is online
+@app.get('/')
+@cache(expire=ONE_WEEK_SEC, namespace='status_check')  # Cache for 1 week
+async def status_check():
+    return {"Status": "API is online..."}
+
+
+@cache(expire=ONE_DAY_SEC, namespace='pipeline_classifier')  # Cache for 1 day
+async def pipeline_classifier(pipeline: imbPipeline, encoder: LabelEncoder, data: SepsisFeatures) -> ErrorResponse | PredictionResponse:
+    output = ErrorResponse(**{'execution_msg': None,
+                              'execution_code': None, 'result': None})
+    try:
+        # Create dataframe
+        df = pd.DataFrame([data.model_dump()])
+
+        # Make prediction
+        prediction = pipeline.predict(df)
+
+        pred_int = int(prediction[0])
+
+        prediction = encoder.inverse_transform([pred_int])[0]
+
+        # Get the probability of the predicted class
+        probability = round(
+            float(pipeline.predict_proba(df)[0][pred_int] * 100), 2)
+
+        msg = 'Execution was successful'
+        code = 1
+        result = {"prediction": prediction, "probability": probability}
+
+        output = PredictionResponse(
+            **{'execution_msg': msg,
+               'execution_code': code, 'result': result}
+        )
+
+    except Exception as e:
+        msg = 'Execution failed'
+        code = 0
+        result = {'error': f"Omg, pipeline classsifier failure{e}"}
+        output = ErrorResponse(**{'execution_msg': msg,
+                                  'execution_code': code, 'result': result})
+
+    finally:
+        return output
+
+
+# Xgboost endpoint: classify sepsis with xgboost
+@app.post('/xgboost_prediction')
+async def xgboost_classifier(data: SepsisFeatures) -> ErrorResponse | PredictionResponse:
+    xgboost_pipeline, encoder = await load_pipeline(XGBOOST_URL, ENCODER_URL)
+    output = await pipeline_classifier(xgboost_pipeline, encoder, data)
+    return output
+
+
+# Random forest endpoint: classify sepsis with random forest
+@app.post('/random_forest_prediction')
+async def random_forest_classifier(data: SepsisFeatures) -> ErrorResponse | PredictionResponse:
+    random_forest_pipeline, encoder = await load_pipeline(RANDOM_FOREST_URL, ENCODER_URL)
+    output = await pipeline_classifier(random_forest_pipeline, encoder, data)
+    return output

src/api/config.py

@@ -0,0 +1,15 @@
+import os
+
+ONE_DAY_SEC = 24*60*60
+
+ONE_WEEK_SEC = ONE_DAY_SEC*7
+
+XGBOOST_URL = "https://raw.githubusercontent.com/valiantezabuku/Building-Machine-Learning-API-s-with-FastAPI-Collabo/develop/dev/models/xgboost.joblib"
+
+RANDOM_FOREST_URL = "https://raw.githubusercontent.com/valiantezabuku/Building-Machine-Learning-API-s-with-FastAPI-Collabo/develop/dev/models/random_forest.joblib"
+
+ENCODER_URL = "https://raw.githubusercontent.com/valiantezabuku/Building-Machine-Learning-API-s-with-FastAPI-Collabo/develop/dev/models/encoder.joblib"
+
+BASE_DIR = './'  # Where Unicorn server runs from
+
+ENV_PATH = os.path.join(BASE_DIR, 'src/api/.env')