Маємо те, що маємо. Усе що відбувається

Показ дописів із міткою SQL. Показати всі дописи

2026-03-29

Comparing Performance After Optimization in django-mariadb-vector

Since version v0.2.0, the 'django-mariadb-vector' library includes several optimization options. Here are the results from performance tests.

Performance results

Note: Performance was measured using 20,000 iterations (3 runs) in `tests/measure_performance.py`, with randomly generated vectors of dimension 3072.

Benefits 'orlson' vs 'json'

Up to ~20× faster compared to the standard `json` library on generate vector data
Up to ~8× faster compared to the standard `json` library on response vector data

Benefits of 'binary' on response vector data

Up to ~16× faster compared to the standard `json` library
About ~2× faster compared to `orjson`

Testing output

Warming for 3 seconds

decode_binary start testing...
binary: 1.8913245666384075 [1.9641265999525785, 1.8629208999918774, 1.846926199970767]

decode_orjson_str start testing...
json_str: 3.804304266697727 [4.095871299970895, 3.7040354000637308, 3.6130061000585556]

decode_json_str start testing...
json_str: 30.897333099972457 [30.78853879997041, 31.119306599954143, 30.784153899992816]
* decode binary is fastest than json in 16.34 times
* decode binary is fastest than orjson in 2.01 times
* decode orjson is fastest than json in 8.12 times

encode_orjson_str start testing...
orjson_str: 2.32788303331472 [2.455615399987437, 2.2925777999917045, 2.235455899965018]

encode_json_str start testing...
json_str: 51.04346506666237 [50.961705199908465, 50.88195970002562, 51.28673030005302]
* encode orjson is fastest than json in 21.93 times

Reference

🔗 DEMO Django application of usage library 'django-mariadb-vector' (MariaDB Vector)

🔗 Django MariaDB Vector package on pypi - 'django-mariadb-vector'

🔗 Repo library: https://github.com/lexxai/django-mariadb-vector

🔗 Repo Demo: https://github.com/lexxai/django-mariadb-vector-demo

2026-03-28

DEMO Django application of usage library 'django-mariadb-vector' (MariaDB Vector)

pip install django-mariadb-vector

📒 Django MariaDB Vector DEMO application

A minimal demo project showing how to build article recommendations using vector similarity in Django with MariaDB as the database.

The app stores articles, embeds their content into vectors, and then finds similar articles based on vector distance.

🔗 Repo: https://github.com/lexxai/django-mariadb-vector-demo
🔗 Examples: https://github.com/lexxai/django-mariadb-vector-demo/tree/main/docs

🔗 Repo library: https://github.com/lexxai/django-mariadb-vector

Features

Django application using MariaDB as the primary database
Article model with text content
Vector-based similarity search for recommendations
Simple UI:
List of all articles

"Similar articles" view for a selected article
Admin interface to add and manage articles

Детальніше »

2024-10-19

FastAPI Dependency Injection: 21 Examples Using Depends() with AI Answers

Why ?

Once I wanted to understand how Depends works in FastAPI more thoroughly to understand what's under the hood. I started communicating with the AI. The answer was a question, and I got hooked. 

I decided to test it in real code and created a training project and wrote this post for it based on my questions to the AI and its answers.
Some answers were wrong, some I had already optimized a bit. But all the results shown are real and created by me personally. 

The result was 21 training examples written in Python, where I checked the answers and suggestions from the AI.

Детальніше »

2023-12-04

Нотатка: Python. Django framework. ORM "GROUP BY". SQLite.

Django SQL GROUP BY.

А Ви знали що його не має в ORM у прямому вигляді ?

SELECT x.* FROM logs x WHERE x.username = 'user1' GROUP BY x.host

DBeaver тестування

Є таблиця log доступу користувачів:

CREATE TABLE logs (
	id INTEGER PRIMARY KEY AUTOINCREMENT,
	date DATETIME,
	host VARCHAR,
	request VARCHAR,
	username VARCHAR
);

CREATE INDEX logs_host_IDX ON logs (host);

SEED LOG TABLE

Стала проста задача для SQL запиту. Для отримання унікальних записів IP адрес з яких отримував доступ певний користувач.

Детальніше »

2023-08-14

Full course : Google Cybersecurity Professional Certificate. Put It to Work: Prepare for Cybersecurity Jobs - Completed on Aug 6, 2023

Ось і 8-й курс "Put It to Work: Prepare for Cybersecurity Jobs" з 8-ми курсів за напрямком Google Cybersecurity Professional Certificate на платформі Coursera та за підтримки (2022-2023 UA Prometheus) - успішно завершено. За що отримано підсумковий сертифікат за всі 8 курсів.