Skip to content

README.md

Latest commit

 

History

History
66 lines (47 loc) · 1.1 KB

README.md

File metadata and controls

66 lines (47 loc) · 1.1 KB
Raw

Interactive Streamlit app with chdb

In this recipe, we'll learn how to create an interactive Streamlit app to analyse energy usage with chdb.

Generate data

Generate usage data

poetry run python datagen.py

This writes data into data.csv

Download Clickhouse

curl https://clickhouse.com/ | sh

Setup ClickHouse

Launch ClickHouse Local

./clickhouse local -m --path energy.chdb

## Ingest data

CREATE DATABASE energy;
CREATE TABLE IF NOT EXISTS energy.tariffs 
ENGINE = MergeTree
ORDER BY day 
AS
SELECT arrayJoin(arrayMap(
    x -> addDays(startDate, x),
    range(dateDiff('days', startDate, endDate) + 1)
)) AS day, *
FROM file(`data/tariffs.csv`)
SETTINGS schema_inference_make_columns_nullable=0;
CREATE TABLE energy.usage
ENGINE = MergeTree
ORDER BY epochTimestamp AS
SELECT *
FROM `data/data.csv`
SETTINGS schema_inference_make_columns_nullable=0;

Launch Streamlit app

Make sure you exit ClickHouse Local first (type exit;)

poetry run streamlit run app.py --server.headless true

Open http://localhost:8501