Create app.py

app.py

@@ -0,0 +1,120 @@
+# app.py  (new UI wrapper)
+
+import os
+from pathlib import Path
+import gradio as gr
+
+from rag_core import (          # this is your current file, renamed
+    rag_reply,
+    W_TFIDF_DEFAULT,
+    W_BM25_DEFAULT,
+    W_EMB_DEFAULT,
+    LOG_PATH,
+    ARTIFACT_DIR,
+)
+from rag_eval_metrics import evaluate_rag
+
+# ------------- RAG chat wrapper ----------------
+def rag_chat_fn(
+    message,
+    history,
+    top_k,
+    n_sentences,
+    include_passages,
+    w_tfidf,
+    w_bm25,
+    w_emb,
+):
+    if not message or not message.strip():
+        return "Ask a literature question (e.g., *How does CNT length affect gauge factor?*)"
+    return rag_reply(
+        question=message,
+        k=int(top_k),
+        n_sentences=int(n_sentences),
+        include_passages=bool(include_passages),
+        use_llm=False,
+        model=None,
+        temperature=0.2,
+        strict_quotes_only=False,
+        w_tfidf=float(w_tfidf),
+        w_bm25=float(w_bm25),
+        w_emb=float(w_emb),
+        config_id=None,
+    )
+
+# ------------- Evaluate wrapper ----------------
+def run_eval_ui(gold_file, k):
+    if gold_file is None:
+        # Assume default gold.csv at repo root
+        gold_path = Path("gold.csv")
+        if not gold_path.exists():
+            return (
+                "**No gold.csv provided or found in the working directory.**\n"
+                "Upload a file or place gold.csv next to app.py."
+            )
+        gold_csv = str(gold_path)
+    else:
+        gold_csv = gold_file.name
+
+    logs_jsonl = str(LOG_PATH)
+    out_dir = str(ARTIFACT_DIR)
+    # This prints to console and writes CSV/JSON; we return a short message for the UI
+    evaluate_rag(gold_csv, logs_jsonl, k=int(k), out_dir=out_dir, group_by_weights=True)
+    return (
+        f"✅ Evaluation finished.\n\n"
+        f"- Per-question metrics: `{ARTIFACT_DIR / 'metrics_per_question.csv'}`\n"
+        f"- Aggregate metrics: `{ARTIFACT_DIR / 'metrics_aggregate.json'}`\n"
+        f"- Config surface: `{ARTIFACT_DIR / 'metrics_by_weights.csv'}`"
+    )
+
+# ------------- Build Gradio UI -----------------
+with gr.Blocks(title="Self-Sensing Concrete RAG") as demo:
+    gr.Markdown(
+        "<h1>Self-Sensing Concrete Assistant — Hybrid RAG</h1>"
+        "<p>Ask questions about self-sensing concrete; answers are grounded in your local PDFs.</p>"
+    )
+
+    with gr.Tabs():
+        # --------- RAG Chat tab ---------
+        with gr.Tab("📚 RAG Chat"):
+            with gr.Row():
+                top_k = gr.Slider(3, 15, value=8, step=1, label="Top-K chunks")
+                n_sentences = gr.Slider(2, 8, value=4, step=1, label="Answer length (sentences)")
+                include_passages = gr.Checkbox(
+                    value=False, label="Include supporting passages"
+                )
+            with gr.Row():
+                w_tfidf = gr.Slider(0.0, 1.0, value=W_TFIDF_DEFAULT, step=0.05, label="TF-IDF weight")
+                w_bm25  = gr.Slider(0.0, 1.0, value=W_BM25_DEFAULT,  step=0.05, label="BM25 weight")
+                w_emb   = gr.Slider(0.0, 1.0, value=W_EMB_DEFAULT,   step=0.05, label="Dense weight")
+
+            gr.ChatInterface(
+                fn=rag_chat_fn,
+                additional_inputs=[top_k, n_sentences, include_passages, w_tfidf, w_bm25, w_emb],
+                title="Hybrid RAG Q&A",
+                description="Hybrid BM25 + TF-IDF + dense retrieval with MMR sentence selection."
+            )
+
+        # --------- Evaluation tab ---------
+        with gr.Tab("📏 Evaluate RAG"):
+            gr.Markdown(
+                "Upload **gold.csv** and compute retrieval metrics against `rag_artifacts/rag_logs.jsonl`."
+            )
+            gold_file = gr.File(label="gold.csv", file_types=[".csv"])
+            k_slider  = gr.Slider(3, 15, value=8, step=1, label="k for Hit/Recall/nDCG")
+            btn_eval  = gr.Button("Run Evaluation")
+            eval_out  = gr.Markdown(label="Evaluation log")
+
+            btn_eval.click(
+                fn=run_eval_ui,
+                inputs=[gold_file, k_slider],
+                outputs=eval_out,
+            )
+
+# ------------- Launch app -----------------
+if __name__ == "__main__":
+    demo.queue().launch(
+        server_name="0.0.0.0",
+        server_port=7860,
+        share=False,
+    )