From 9864a2b551c6772f0d0c2fe81b0bff9301f23776 Mon Sep 17 00:00:00 2001
From: Debanjum Singh Solanky <debanjum@gmail.com>
Date: Sat, 31 Jul 2021 00:20:37 -0700
Subject: [PATCH] Retrieve most relevant entries for a query using MSMarco
 based bi-encoder

Returns best 3 results ranked by MSMarco based biencoder score of
query match to entries from org-mode notes
---
 asymmetric.py | 71 +++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 71 insertions(+)
 create mode 100644 asymmetric.py

diff --git a/asymmetric.py b/asymmetric.py
new file mode 100644
index 00000000..2e84bc20
--- /dev/null
+++ b/asymmetric.py
@@ -0,0 +1,71 @@
+#!/usr/bin/env python
+
+import json
+from sentence_transformers import SentenceTransformer, CrossEncoder, util
+import time
+import gzip
+import os
+import sys
+
+# We use the Bi-Encoder to encode all passages, so that we can use it with sematic search
+model_name = 'msmarco-MiniLM-L-6-v3'
+bi_encoder = SentenceTransformer(model_name)
+top_k = 100  # Number of passages we want to retrieve with the bi-encoder
+
+# The bi-encoder will retrieve 100 documents.
+# We use a cross-encoder, to re-rank the results list to improve the quality
+cross_encoder = CrossEncoder('cross-encoder/ms-marco-MiniLM-L-6-v2')
+
+# We split these articles into paragraphs and encode them with the bi-encoder
+notes_filepath = 'Notes.jsonl.gz'
+
+passages = []
+with gzip.open(notes_filepath, 'rt', encoding='utf8') as fIn:
+    for line in fIn:
+        data = json.loads(line.strip())
+        passages.extend([f'{data["Title"]}\n{data["Body"] if "Body" in data else ""}'])
+
+print(f"Passages: {len(passages)}")
+
+# Here, we compute the corpus_embeddings from scratch (which can take a while depending on the GPU)
+corpus_embeddings = bi_encoder.encode(passages, convert_to_tensor=True, show_progress_bar=True)
+
+# This function will search all notes for passages that answer the query
+def search(query):
+    print("Input question:", query)
+
+    ##### Sematic Search #####
+    # Encode the query using the bi-encoder and find potentially relevant passages
+    question_embedding = bi_encoder.encode(query, convert_to_tensor=True)
+    #question_embedding = question_embedding.cuda()
+    hits = util.semantic_search(question_embedding, corpus_embeddings, top_k=top_k)
+    hits = hits[0]  # Get the hits for the first query
+
+    ##### Re-Ranking #####
+    ## Now, score all retrieved passages with the cross_encoder
+    #cross_inp = [[query, passages[hit['corpus_id']]] for hit in hits]
+    #cross_scores = cross_encoder.predict(cross_inp)
+    #
+    ## Sort results by the cross-encoder scores
+    #for idx in range(len(cross_scores)):
+    #    hits[idx]['cross-score'] = cross_scores[idx]
+
+    # Output of top-5 hits from bi-encoder
+    print("\n-------------------------\n")
+    print("Top-3 Bi-Encoder Retrieval hits")
+    hits = sorted(hits, key=lambda x: x['score'], reverse=True)
+    for hit in hits[0:3]:
+        print("\t{:.3f}\t{}".format(hit['score'], passages[hit['corpus_id']].replace("\n", " ")))
+
+    # Output of top-5 hits from re-ranker
+    #print("\n-------------------------\n")
+    #print("Top-3 Cross-Encoder Re-ranker hits")
+    #hits = sorted(hits, key=lambda x: x['cross-score'], reverse=True)
+    #for hit in hits[0:3]:
+    #    print("\t{:.3f}\t{}".format(hit['cross-score'], passages[hit['corpus_id']].replace("\n", " ")))
+
+while True:
+    user_query = input("Enter your query: ")
+    if user_query == "exit":
+        exit(0)
+    search(query = user_query)