Wrap asymmetric search model into SearchModels. Test notes search end-to-end

- Wrap asymmetric search model parameters into AsymmetricSearchModel class
- Create wrapper for all search type models. Put notes search model into it
- Test notes search end-to-end from client API layer to results.
  Use model build on test data
This commit is contained in:
Debanjum Singh Solanky 2021-09-29 20:24:27 -07:00
parent cde11a2331
commit e22e0b41e3
4 changed files with 51 additions and 27 deletions

View file

@ -11,9 +11,11 @@ from fastapi import FastAPI
from search_type import asymmetric, symmetric_ledger, image_search
from utils.helpers import get_from_dict
from utils.cli import cli
from utils.config import SearchType, SearchSettings
from utils.config import SearchType, SearchSettings, SearchModels
# Application Global State
model = SearchModels()
search_settings = SearchSettings()
app = FastAPI()
@ -29,16 +31,10 @@ def search(q: str, n: Optional[int] = 5, t: Optional[SearchType] = None):
if (t == SearchType.Notes or t == None) and search_settings.notes_search_enabled:
# query notes
hits = asymmetric.query_notes(
user_query,
corpus_embeddings,
entries,
bi_encoder,
cross_encoder,
top_k)
hits = asymmetric.query_notes(user_query, model.notes_search)
# collate and return results
return asymmetric.collate_results(hits, entries, results_count)
return asymmetric.collate_results(hits, model.notes_search.entries, results_count)
if (t == SearchType.Music or t == None) and search_settings.music_search_enabled:
# query music library
@ -90,9 +86,7 @@ def search(q: str, n: Optional[int] = 5, t: Optional[SearchType] = None):
def regenerate(t: Optional[SearchType] = None):
if (t == SearchType.Notes or t == None) and search_settings.notes_search_enabled:
# Extract Entries, Generate Embeddings
global corpus_embeddings
global entries
entries, corpus_embeddings, _, _, _ = asymmetric.setup(
models.notes_search = asymmetric.setup(
org_config['input-files'],
org_config['input-filter'],
pathlib.Path(org_config['compressed-jsonl']),
@ -146,7 +140,7 @@ if __name__ == '__main__':
org_config = get_from_dict(args.config, 'content-type', 'org')
if org_config and ('input-files' in org_config or 'input-filter' in org_config):
search_settings.notes_search_enabled = True
entries, corpus_embeddings, bi_encoder, cross_encoder, top_k = asymmetric.setup(
model.notes_search = asymmetric.setup(
org_config['input-files'],
org_config['input-filter'],
pathlib.Path(org_config['compressed-jsonl']),

View file

@ -17,6 +17,7 @@ from sentence_transformers import SentenceTransformer, CrossEncoder, util
# Internal Packages
from utils.helpers import get_absolute_path, resolve_absolute_path
from processor.org_mode.org_to_jsonl import org_to_jsonl
from utils.config import AsymmetricSearchModel
def initialize_model():
@ -64,7 +65,7 @@ def compute_embeddings(entries, bi_encoder, embeddings_file, regenerate=False, v
return corpus_embeddings
def query_notes(raw_query, corpus_embeddings, entries, bi_encoder, cross_encoder, top_k=100):
def query_notes(raw_query: str, model: AsymmetricSearchModel):
"Search all notes for entries that answer the query"
# Separate natural query from explicit required, blocked words filters
query = " ".join([word for word in raw_query.split() if not word.startswith("+") and not word.startswith("-")])
@ -72,20 +73,22 @@ def query_notes(raw_query, corpus_embeddings, entries, bi_encoder, cross_encoder
blocked_words = set([word[1:].lower() for word in raw_query.split() if word.startswith("-")])
# Encode the query using the bi-encoder
question_embedding = bi_encoder.encode(query, convert_to_tensor=True)
question_embedding = model.bi_encoder.encode(query, convert_to_tensor=True)
# Find relevant entries for the query
hits = util.semantic_search(question_embedding, corpus_embeddings, top_k=top_k)
hits = util.semantic_search(question_embedding, model.corpus_embeddings, top_k=model.top_k)
hits = hits[0] # Get the hits for the first query
# Filter results using explicit filters
hits = explicit_filter(hits, [entry[0] for entry in entries], required_words, blocked_words)
hits = explicit_filter(hits,
[entry[0] for entry in model.entries],
required_words,blocked_words)
if hits is None or len(hits) == 0:
return hits
# Score all retrieved entries using the cross-encoder
cross_inp = [[query, entries[hit['corpus_id']][0]] for hit in hits]
cross_scores = cross_encoder.predict(cross_inp)
cross_inp = [[query, model.entries[hit['corpus_id']][0]] for hit in hits]
cross_scores = model.cross_encoder.predict(cross_inp)
# Store cross-encoder scores in results dictionary for ranking
for idx in range(len(cross_scores)):
@ -161,7 +164,7 @@ def setup(input_files, input_filter, compressed_jsonl, embeddings, regenerate=Fa
# Compute or Load Embeddings
corpus_embeddings = compute_embeddings(entries, bi_encoder, embeddings, regenerate=regenerate, verbose=verbose)
return entries, corpus_embeddings, bi_encoder, cross_encoder, top_k
return AsymmetricSearchModel(entries, corpus_embeddings, bi_encoder, cross_encoder, top_k)
if __name__ == '__main__':

View file

@ -6,7 +6,7 @@ import pytest
from fastapi.testclient import TestClient
# Internal Packages
from main import app
from main import app, search_settings, model
from search_type import asymmetric
@ -55,18 +55,33 @@ def test_regenerate_with_valid_search_type():
# ----------------------------------------------------------------------------------------------------
def test_asymmetric_setup():
def test_notes_search():
# Arrange
input_files = [Path('tests/data/main_readme.org'), Path('tests/data/interface_emacs_readme.org')]
input_filter = None
compressed_jsonl = Path('tests/data/.test.jsonl.gz')
embeddings = Path('tests/data/.test_embeddings.pt')
regenerate = False
verbose = 1
# Act
entries, corpus_embeddings, bi_encoder, cross_encoder, top_k = asymmetric.setup(input_files, input_filter, compressed_jsonl, embeddings, regenerate, verbose)
# Regenerate embeddings during asymmetric setup
notes_model = asymmetric.setup(input_files, input_filter, compressed_jsonl, embeddings, regenerate=True, verbose=0)
# Assert
assert len(entries) == 10
assert len(corpus_embeddings) == 10
assert len(notes_model.entries) == 10
assert len(notes_model.corpus_embeddings) == 10
# Arrange
model.notes_search = notes_model
search_settings.notes_search_enabled = True
user_query = "How to call semantic search from Emacs?"
# Act
response = client.get(f"/search?q={user_query}&n=1&t=notes")
# Assert
assert response.status_code == 200
# assert actual_data contains "Semantic Search via Emacs"
search_result = response.json()[0]["Entry"]
assert "Semantic Search via Emacs" in search_result

View file

@ -18,3 +18,15 @@ class SearchSettings():
image_search_enabled: bool = False
class AsymmetricSearchModel():
def __init__(self, entries, corpus_embeddings, bi_encoder, cross_encoder, top_k):
self.entries = entries
self.corpus_embeddings = corpus_embeddings
self.bi_encoder = bi_encoder
self.cross_encoder = cross_encoder
self.top_k = top_k
@dataclass
class SearchModels():
notes_search: AsymmetricSearchModel = None