Do I need specific libraries installed to use this skill?

The generated scripts require pandas; additional support for Parquet or SQL requires pyarrow or sqlalchemy respectively.

What is the standard Nixtla schema required by this skill?

Nixtla models typically require three main columns: 'unique_id' for the series identifier, 'ds' for the timestamp, and 'y' for the target numeric value.

Does this skill generate executable code?

Yes, it generates Python transformation modules that you can run directly to convert your raw data into the validated Nixtla format.

How does the skill handle ambiguous column names?

It uses intelligent inference to detect the most likely candidates for time and target columns, but provides clear error handling and manual override options if multiple candidates exist.

Which data formats does the Nixtla Schema Mapper support?

The skill supports a wide range of formats including CSV, Parquet files, SQL tables/queries via SQLAlchemy, and dbt models.

Nixtla Schema Mapper

Name: Nixtla Schema Mapper
Author: intent-solutions-io

byintent-solutions-io

数据科学与机器学习

Transforms diverse data sources into Nixtla-compatible formats for time-series forecasting and analysis.

关于

The Nixtla Schema Mapper skill automates the critical and often tedious process of reformatting raw data into the standardized schema required by Nixtla models (unique_id, ds, y). It intelligently infers column roles—identifying timestamps, targets, and series IDs—across various sources including CSV, SQL, and dbt models. By generating production-ready Python transformation scripts and schema contracts, it ensures data quality and reproducibility, making it an essential tool for developers building forecasting pipelines with TimeGPT or StatsForecast.

主要功能

Support for complex date parsing and custom column mapping
0 GitHub stars
Python code generation for CSV, SQL, Parquet, and dbt transformations
Built-in data quality checks to prevent forecasting errors
Automated inference of timestamp, target, and series ID columns
Comprehensive schema contract generation with validation rules

使用场景

Standardizing disparate data sources for large-scale model benchmarking
Automating ETL workflows for multi-series time-series data sets
Preparing legacy SQL databases for TimeGPT forecasting pipelines

关于

主要功能

Support for complex date parsing and custom column mapping
0 GitHub stars
Python code generation for CSV, SQL, Parquet, and dbt transformations
Built-in data quality checks to prevent forecasting errors
Automated inference of timestamp, target, and series ID columns
Comprehensive schema contract generation with validation rules

使用场景

Standardizing disparate data sources for large-scale model benchmarking
Automating ETL workflows for multi-series time-series data sets
Preparing legacy SQL databases for TimeGPT forecasting pipelines