DataStore Quickstart
Get up and running with DataStore in minutes. This guide covers installation, migrating from pandas, and basic usage patterns.
Installation
Install chDB with pip:
For optional dependencies:
Verify Installation
One-Line Migration from Pandas
The simplest way to start using DataStore is to change your import statement:
That's it! Your existing pandas code will now use DataStore and benefit from SQL optimization.
Migration Example
Basic Usage
Creating a DataStore
Filtering Data
Selecting Columns
Sorting
Grouping and Aggregation
Joining DataStores
Getting Results
DataStore uses lazy evaluation - operations are not executed until you need the results.
Triggering Execution
Viewing Generated SQL
Output:
Working with Different Data Sources
Local Files
Cloud Storage
Databases
String and DateTime Operations
String Operations
DateTime Operations
ClickHouse Extensions
Best Practices
1. Use Parquet for Large Files
2. Filter Early
3. Select Only Needed Columns
4. Use SQL for Complex Operations
Next Steps
- Learn about all Factory Methods for creating DataStore
- Explore Query Building for SQL-style operations
- Check out Accessors for string, datetime, and more
- Read the Performance Guide for optimization tips