Implementation:Explodinggradients Ragas BaseBackend Class

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

Overview

Abstract base class defining the contract all storage backends must implement for persisting datasets and experiments.

Description

BaseBackend is an abstract class from the Ragas backends module that establishes the standard interface for storage backends. All concrete backends (InMemory, CSV, JSONL, Google Drive) must implement this contract. It defines six abstract methods for loading, saving, and listing both datasets and experiments.

Usage

Extend this class when creating a custom storage backend for Ragas. Implement all six abstract methods to provide persistent storage for datasets and experiments.

Code Reference

Source Location

• Repository: Explodinggradients_Ragas
• File: src/ragas/backends/base.py
• Lines: 9-153

Signature

class BaseBackend(ABC):
    @abstractmethod
    def load_dataset(self, name: str) -> List[Dict[str, Any]]:
        ...
    @abstractmethod
    def load_experiment(self, name: str) -> List[Dict[str, Any]]:
        ...
    @abstractmethod
    def save_dataset(self, name: str, data: List[Dict[str, Any]], data_model: Optional[Type[BaseModel]] = None) -> None:
        ...
    @abstractmethod
    def save_experiment(self, name: str, data: List[Dict[str, Any]], data_model: Optional[Type[BaseModel]] = None) -> None:
        ...
    @abstractmethod
    def list_datasets(self) -> List[str]:
        ...
    @abstractmethod
    def list_experiments(self) -> List[str]:
        ...

Import

from ragas.backends.base import BaseBackend

I/O Contract

Inputs

Name	Type	Required	Description
name	str	Yes	Name identifier for the dataset or experiment
data	List[Dict[str, Any]]	Yes	List of records to save
data_model	Optional[Type[BaseModel]]	No	Pydantic model for validation context

Outputs

Name	Type	Description
load_dataset returns	List[Dict[str, Any]]	Loaded dataset records
load_experiment returns	List[Dict[str, Any]]	Loaded experiment records
list_datasets returns	List[str]	Sorted list of dataset names
list_experiments returns	List[str]	Sorted list of experiment names

Usage Examples

Custom Backend Implementation

from ragas.backends.base import BaseBackend
from typing import Any, Dict, List, Optional, Type
from pydantic import BaseModel

class MyCustomBackend(BaseBackend):
    def __init__(self):
        self._store = {}

    def load_dataset(self, name: str) -> List[Dict[str, Any]]:
        if name not in self._store:
            raise FileNotFoundError(f"Dataset '{name}' not found")
        return self._store[name]

    def save_dataset(self, name: str, data: List[Dict[str, Any]], data_model: Optional[Type[BaseModel]] = None) -> None:
        self._store[name] = data

    def load_experiment(self, name: str) -> List[Dict[str, Any]]:
        raise FileNotFoundError(f"Not implemented")

    def save_experiment(self, name: str, data: List[Dict[str, Any]], data_model: Optional[Type[BaseModel]] = None) -> None:
        pass

    def list_datasets(self) -> List[str]:
        return sorted(self._store.keys())

    def list_experiments(self) -> List[str]:
        return []