readme update

README.md

@@ -0,0 +1,21 @@
+## Index details
+
+This dataset contains a FAISS index and associated chunk metadata to support
+retrieval-augmented generation (RAG) use cases on [hkir-dev/ndl-core-corpus](https://huggingface.co/datasets/hkir-dev/ndl-core-corpus).
+
+### Embeddings
+- Model: sentence-transformers/all-MiniLM-L6-v2
+- Dimension: 384
+- Normalisation: L2
+- Similarity: cosine (inner product)
+
+### Chunking
+- Strategy: recursive character-based chunking
+- Chunk size: 800 characters
+- Overlap: 100 characters
+
+### Source data
+Chunks reference records in the dataset:
+[hkir-dev/ndl-core-corpus](https://huggingface.co/datasets/hkir-dev/ndl-core-corpus)
+
+See `rag_config.json` for full, machine-readable configuration.

rag_config.json

@@ -0,0 +1,25 @@
+{
+  "index_type": "faiss",
+  "faiss_index": {
+    "type": "IndexFlatIP",
+    "metric": "cosine"
+  },
+  "embedding": {
+    "model_name": "sentence-transformers/all-MiniLM-L6-v2",
+    "provider": "sentence-transformers",
+    "dimension": 384,
+    "normalised": true
+  },
+  "chunking": {
+    "strategy": "recursive",
+    "unit": "characters",
+    "chunk_size": 800,
+    "chunk_overlap": 100
+  },
+  "source": {
+    "dataset": "hkir-dev/ndl-core-corpus",
+    "id_field": "identifier"
+  },
+  "created_at": "2025-01-01",
+  "version": "v0.1"
+}