Create app.py

app.py

@@ -0,0 +1,38 @@
+from fastapi import FastAPI, Request
+from fastapi.responses import JSONResponse
+from bertopic import BERTopic
+from sklearn.feature_extraction.text import CountVectorizer
+from sklearn.decomposition import TruncatedSVD
+from sklearn.cluster import KMeans
+import pandas as pd
+
+vectorizer_model = CountVectorizer()
+dimensionality_model = TruncatedSVD(n_components=5)
+clustering_model = KMeans(n_clusters=5, random_state=42)
+
+topic_model = BERTopic(
+    vectorizer_model=vectorizer_model,
+    umap_model=dimensionality_model,
+    hdbscan_model=clustering_model
+)
+
+app = FastAPI()
+
+@app.post("/predict")
+async def predict(request: Request):
+    data = await request.json()
+    if "text" in data:
+        text = data["text"]
+    elif "data" in data and isinstance(data["data"], list):
+        text = data["data"][0]
+    else:
+        return JSONResponse({"error": "No input text provided."}, status_code=400)
+    documents = [doc.strip() for doc in text.split("\n") if doc.strip()]
+    if not documents:
+        return JSONResponse({"error": "No valid input."}, status_code=400)
+    topics, probs = topic_model.fit_transform(documents)
+    topic_info = topic_model.get_topic_info()
+    return {
+        "topics": topic_info.to_dict(orient="records"),
+        "topic_assignments": topics
+    }