first commit

Dockerfile

@@ -0,0 +1,13 @@
+FROM python:3.9
+
+RUN useradd -m -u 1000 user
+USER user
+ENV PATH="/home/user/.local/bin:$PATH"
+
+WORKDIR /app
+
+COPY --chown=user ./requirements.txt requirements.txt
+RUN pip install --no-cache-dir --upgrade -r requirements.txt
+
+COPY --chown=user . /app
+CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]

app.py

@@ -0,0 +1,34 @@
+from fastapi import FastAPI
+import lancedb
+from sentence_transformers import SentenceTransformer
+from huggingface_hub import snapshot_download
+
+app = FastAPI()
+
+# 1. Download ONLY the LanceDB folder (Saved space/time by ignoring FAISS)
+print("⏳ Downloading LanceDB index...")
+index_path = snapshot_download(
+    repo_id="hkir-dev/ndl-core-rag-index",
+    repo_type="dataset",
+    allow_patterns="lancedb_search_index/*"  # only need this folder, not the FAISS one
+)
+
+# 2. Connect DB and load model
+db = lancedb.connect(f"{index_path}/lancedb_search_index")
+table = db.open_table("ndl_core_datasets")
+all_columns = table.schema.names
+columns_to_select = [col for col in all_columns if col != "vector"]
+
+model = SentenceTransformer('all-MiniLM-L6-v2')
+
+@app.get("/search")
+def search(query: str, limit: int = 5):
+    query_vector = model.encode(query)
+    results = (
+        table.search(query_vector)      # Your vector search
+        .metric("cosine")               # Ensure metric matches your index
+        .select(columns_to_select)      # <--- The key step: explicit column selection
+        .limit(5)                       # Number of results
+        .to_pandas()                    # Convert to DataFrame
+    )
+    return results.to_dict(orient='records')

requirements.txt

@@ -0,0 +1,9 @@
+fastapi
+uvicorn
+lancedb
+sentence-transformers
+pandas
+huggingface-hub
+pyarrow
+torch
+numpy