Enhance FastAPI application with text preprocessing and update Dockerfile for pip installation

Dockerfile

@@ -40,7 +40,8 @@ ENV TRANSFORMERS_CACHE=/app/.cache/huggingface/hub
 # Leverage a cache mount to /root/.cache/pip to speed up subsequent builds.
 # Leverage a bind mount to requirements.txt to avoid having to copy them into
 # into this layer.
-RUN python -m pip install -r requirements.txt
+RUN python -m pip install --upgrade pip \
+    && python -m pip install --no-cache-dir -r requirements.txt
 
 # Switch to the non-privileged user to run the application.
 USER appuser

app.py

@@ -1,17 +1,39 @@
 from fastapi import FastAPI
 from transformers import pipeline
+import re
 
 pipe = pipeline("text-classification", model="JungleLee/bert-toxic-comment-classification")
 
-app = FastAPI()
+app = FastAPI(
+    title="Hopeline - AI Inference API",
+    description="API for detecting toxic comments",
+    version="0.1"
+)
+
+def preprocess_text(text: str) -> str:
+    # Remove special characters and extra whitespace
+    text = re.sub(r'[^\w\s]', '', text)
+    # Convert to lowercase
+    text = text.lower()
+    # Remove extra whitespaces
+    text = ' '.join(text.split())
+    return text
 
 @app.get("/")
 async def welcome():
     return "Welcome to Hopeline - AI Inference API"
 
-@app.get('/predict')
-async def predict(text: str):
-    prediction = pipe(text)
+@app.post('/predict')
+async def predict_post(request_body: dict):
+    text = request_body.get('text', '')
+    if not text:
+        return {"error": "No text provided"}
+    
+    # Preprocess text
+    processed_text = preprocess_text(text)
+    
+    # Get prediction
+    prediction = pipe(processed_text)
     return prediction
 
 

requirements.txt

@@ -2,4 +2,4 @@ fastapi
 uvicorn[standard]
 transformers
 torch
-torchvision
+torchvision