> ## Documentation Index
> Fetch the complete documentation index at: https://docs.athina.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Running Evals

<Tip>
  The easiest way to get started is to follow this
  [notebook](https://github.com/athina-ai/athina-evals/blob/main/examples/run_experiment.ipynb)
</Tip>

#### 1. Configure API Keys[](#1-configure-api-keys)

Evals use OpenAI, so you need to configure your OpenAI API key.

If you wish to view the results on [Athina Develop](https://docs.athina.ai/develop), and maintain a historical record of prompts and experiments you run during your development workflow, then you also need an Athina API Key.

```python
from athina.keys import AthinaApiKey, OpenAiApiKey

OpenAiApiKey.set_key(os.getenv('OPENAI_API_KEY'))
AthinaApiKey.set_key(os.getenv('ATHINA_API_KEY')) # optional
```

#### 2. Load your dataset[](#2-load-your-dataset)

[Loading a dataset](api-reference/evals/loading-data/loading-data-for-eval) is quite straightforward - we support JSON and CSV formats.

```python
from athina.loaders import Loader

# Load the data from JSON, Athina or Dictionary
dataset = Loader().load_json(json_file)
```

#### 3. Run an eval on a dataset[](#3-run-an-eval-on-a-dataset)

Running evals on a batch of datapoints is the most effective way to rapidly iterate as you're developing your model.

```python
from athina.evals import ContextContainsEnoughInformation

# Run the ContextContainsEnoughInformation evaluator on the dataset
ContextContainsEnoughInformation(
    model="gpt-4-1106-preview",
    max_parallel_evals=5, # optional, speeds up evals
).run_batch(dataset).to_df()
```

Your results will be printed out as a dataframe that looks like this.

![](https://mintlify.s3.us-west-1.amazonaws.com/athinaai/images/run-single-eval1.png)

***

#### How do I know which fields I need in my dataset?[](#how-do-i-know-which-fields-i-need-in-my-dataset)

<Info>
  For the [RAG Evals](api-reference/evals/preset-evals/rag), we need 3 fields: query, context, and response.

  For these evals, you should use the [RagLoader](https://github.com/athina-ai/athina-evals/blob/main/athina/loaders/rag_loader.py) to load your data. This will ensure the data is in the right format for evals.
</Info>

Every evaluator has a `REQUIRED_ARGS` property that defines the parameters it expects.

If you pass the wrong parameters, the evaluator will raise a `ValueError` telling you what params you are missing.

For example:, the [`Faithfulness` ](https://github.com/athina-ai/athina-evals/blob/main/athina/evals/llm/faithfulness/evaluator.py) evaluator expects `response` and `context` fields.

***

#### Run an eval on a single datapoint[](#run-an-eval-on-a-single-datapoint)

Running an eval on a single datapoint is very simple.

This might be useful if you are trying to run the eval immediately after inference.

```python
# Run the answer relevance evaluator
# Checks if the LLM response answers the user query sufficiently
DoesResponseAnswerQuery().run(query=query, response=response)
```

Here's a [notebook ](https://github.com/athina-ai/athina-evals/blob/main/examples/run_one.ipynb) you can use to get started.