Updated output predictions and README

README.md

-# PiRADS API
+# BiRADS API
+
+## Compatibilidad
+Esta API ha sido desarrollada y probada con la versión de Python 3.9.7.
 
 
 ## Guía de instalación
 ## Guía de instalación
 1. Configuración:
 1. Configuración:
@@ -7,5 +10,5 @@
 2. Crear un entorno virtual: `python3 -m venv /path/to/environment`
 2. Crear un entorno virtual: `python3 -m venv /path/to/environment`
 3. Activar el entorno virtual: `source /path/to/environment/bin/activate`
 3. Activar el entorno virtual: `source /path/to/environment/bin/activate`
 4. Instalar las bibliotecas necesarias: `pip install -r /path/to/requirements.txt`
 4. Instalar las bibliotecas necesarias: `pip install -r /path/to/requirements.txt`
-5. Lanzar el servicio: `uvicorn main:app --host <HOST> --port <PORT>`
-6. Probar que funciona en http://localhost:PORT/docs 
+5. Lanzar el servicio: `python main.py [-h] [--host HOST] [--port PORT]`
+6. Probar que funciona accediendo (por defecto) a: http://localhost:8000/docs
\ No newline at end of file

app.py

@@ -14,8 +14,9 @@ database.Base.metadata.create_all(bind=database.engine)
 
 
 # APP
 # APP
 app = FastAPI()
 app = FastAPI()
+#app = FastAPI(openapi_url=None) # Disable interactive docs
 
 
-# CORS middleware
+# Allow all CORS middleware
 app.add_middleware(
 app.add_middleware(
     CORSMiddleware,
     CORSMiddleware,
     allow_origins=["*"],
     allow_origins=["*"],
@@ -41,8 +42,16 @@ def _predict(text, model, vectorizer):
     X_test = vectorizer.transform([clean_text])
     X_test = vectorizer.transform([clean_text])
 
 
     # Predict outputs
     # Predict outputs
+    labels = [f"PR{i}" for i in range(7)]
     probs = model.predict_proba(X_test)
     probs = model.predict_proba(X_test)
     probs = [prob[0][1] for prob in probs]
     probs = [prob[0][1] for prob in probs]
+    probs = [
+        {
+            "type": label, 
+            "percentage": f"{prob*100:.2f}%"
+        } for label, prob in zip(labels, probs)
+    ]
+
     return probs
     return probs
 
 
 # Methods
 # Methods
@@ -50,7 +59,7 @@ def _predict(text, model, vectorizer):
 def register(user: schemas.UserCreate, db: Session = Depends(dependencies.get_db)):
 def register(user: schemas.UserCreate, db: Session = Depends(dependencies.get_db)):
     db_user = crud.get_user(db, username=user.username)
     db_user = crud.get_user(db, username=user.username)
     if db_user:
     if db_user:
-        raise HTTPException(status_code=400, detail="Email already registered")
+        raise HTTPException(status_code=400, detail="Username already registered")
     return crud.create_user(db=db, user=user)
     return crud.create_user(db=db, user=user)
 
 
 @app.post("/token", response_model=schemas.Token)
 @app.post("/token", response_model=schemas.Token)

main.py

 import uvicorn
 import uvicorn
+import argparse
 from app import app
 from app import app
 
 
 if __name__ == '__main__':
 if __name__ == '__main__':
-    uvicorn.run(app, host='0.0.0.0', port=1447, log_level='info')
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--host", type=str, required=False, default="0.0.0.0")
+    parser.add_argument("--port", type=int, required=False, default=8000)
+    args = parser.parse_args()
+
+    uvicorn.run(app, host=args.host, port=args.port, log_level='info')

requirements.txt

@@ -30,7 +30,7 @@ python-multipart==0.0.5
 PyYAML==6.0
 PyYAML==6.0
 rfc3986==1.5.0
 rfc3986==1.5.0
 rsa==4.9
 rsa==4.9
-scikit-learn==1.2.1
+scikit-learn==1.0.2
 scipy==1.10.0
 scipy==1.10.0
 six==1.16.0
 six==1.16.0
 sniffio==1.3.0
 sniffio==1.3.0

schemas.py

-from typing import List, Union
+from typing import List, Union, Dict
 
 
 from pydantic import BaseModel
 from pydantic import BaseModel
 
 
@@ -33,4 +33,4 @@ class InputData(BaseModel):
 
 
 class OutputData(BaseModel):
 class OutputData(BaseModel):
     id: int
     id: int
-    prediction: List[float]
+    prediction: List[Dict[str, str]]
\ No newline at end of file