Movatterモバイル変換


[0]ホーム

URL:


Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

Sign up
Appearance settings

Commitb9d1498

Browse files
committed
query example using pdf as input
1 parentd157f34 commitb9d1498

File tree

4 files changed

+85
-1
lines changed

4 files changed

+85
-1
lines changed
924 KB
Binary file not shown.
Lines changed: 61 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,61 @@
1+
frompgmlimportDatabase
2+
importos
3+
fromdatasetsimportload_dataset
4+
fromtimeimporttime
5+
fromdotenvimportload_dotenv
6+
fromrich.consoleimportConsole
7+
frompypdfimportPdfReader
8+
frompgml.dbutilsimportrun_select_statement
9+
10+
load_dotenv()
11+
console=Console()
12+
13+
local_pgml="postgres://postgres@127.0.0.1:5433/pgml_development"
14+
15+
conninfo=os.environ.get("PGML_CONNECTION",local_pgml)
16+
db=Database(conninfo)
17+
18+
collection_name="pdf_collection"
19+
collection=db.create_or_get_collection(collection_name)
20+
21+
filename="lincoln.pdf"
22+
reader=PdfReader("lincoln.pdf")
23+
number_of_pages=len(reader.pages)
24+
documents= []
25+
forpage_number,pageinenumerate(reader.pages):
26+
documents.append({"text":page.extract_text(),"page":page_number,"source":filename})
27+
28+
collection.upsert_documents(documents)
29+
collection.generate_chunks()
30+
collection.generate_embeddings()
31+
32+
start=time()
33+
query="When was Lincoln born?"
34+
results=collection.vector_search(query,top_k=1)
35+
36+
# db.archive_collection(collection_name)
37+
conn=db.pool.getconn()
38+
context=" ".join(results[0]["chunk"].strip().split())
39+
context=context.replace('"','\\"').replace("'","''")
40+
41+
select_statement="""SELECT pgml.transform(
42+
'question-answering',
43+
inputs => ARRAY[
44+
'{
45+
\"question\":\"%s\",
46+
\"context\":\"%s\"
47+
}'
48+
]
49+
) AS answer;"""% (
50+
query,
51+
context,
52+
)
53+
54+
results=run_select_statement(conn,select_statement)
55+
db.pool.putconn(conn)
56+
57+
console.print("\nResults for query '%s'"%query)
58+
console.print(results)
59+
db.archive_collection(collection_name)
60+
_end=time()
61+
console.print("Query time = %0.3f"% (_end-start))

‎pgml-sdks/python/pgml/poetry.lock‎

Lines changed: 23 additions & 1 deletion
Some generated files are not rendered by default. Learn more aboutcustomizing how changed files appear on GitHub.

‎pgml-sdks/python/pgml/pyproject.toml‎

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -21,6 +21,7 @@ ipywidgets = "^8.0.6"
2121
datasets ="^2.12.0"
2222
python-dotenv ="^1.0.0"
2323
pypika ="^0.48.9"
24+
pypdf ="^3.9.1"
2425

2526

2627
[tool.poetry.group.dev.dependencies]

0 commit comments

Comments
 (0)

[8]ページ先頭

©2009-2025 Movatter.jp