Implementation:Explodinggradients Ragas LocalCSVBackend Class

Knowledge Sources	Explodinggradients_Ragas
Domains	Storage, Backend
Last Updated	2026-02-10 00:00 GMT

Overview

File-based storage backend using CSV format for local persistence of tabular datasets and experiments.

Description

LocalCSVBackend implements BaseBackend to store data as CSV files on the local filesystem. Datasets and experiments are saved to separate subdirectories under a configurable root directory. This backend is best suited for simple tabular data without deeply nested structures.

Usage

Use this backend when you need persistent local storage in a human-readable CSV format and your data is primarily flat/tabular.

Code Reference

Source Location

• Repository: Explodinggradients_Ragas
• File: src/ragas/backends/local_csv.py
• Lines: 12-146

Signature

class LocalCSVBackend(BaseBackend):
    def __init__(self, root_dir: str) -> None:
        ...

Import

from ragas.backends.local_csv import LocalCSVBackend

I/O Contract

Inputs

Name	Type	Required	Description
root_dir	str	Yes	Root directory for CSV storage
name	str	Yes	Name of the dataset or experiment
data	List[Dict[str, Any]]	Yes	Records to save as CSV rows

Outputs

Name	Type	Description
load returns	List[Dict[str, Any]]	Records loaded from CSV (all values as strings)
list returns	List[str]	Sorted list of CSV file names (without extension)

Usage Examples

from ragas.backends.local_csv import LocalCSVBackend

backend = LocalCSVBackend(root_dir="./my_data")

# Save a dataset
backend.save_dataset("eval_results", [
    {"question": "What is AI?", "score": "0.85"},
    {"question": "What is ML?", "score": "0.92"},
])

# Load the dataset
data = backend.load_dataset("eval_results")
print(data)

# List available datasets
print(backend.list_datasets())  # ["eval_results"]