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

# runExperiment

***

# Function: runExperiment()

```ts theme={null}
function runExperiment<T>(
  params: RunExperimentParams<T>,
): Promise<RunExperimentOutput>;
```

Defined in: [src/utils/experiments.ts](https://github.com/rungalileo/galileo-js/blob/main/src/utils/experiments.ts)

Runs an experiment by processing each row of a dataset through a specified function.
If metrics are provided, they will be used to evaluate the experiment.

Usage:

```typescript theme={null}
// Run an experiment with a runner function
const results = await runExperiment({
  name: 'my-experiment',
  dataset: [{ country: 'France'}],
  function: async (input) => {
       const response = await openai.chat.completions.create({
         model: 'gpt-4o-mini',
         messages: [
           {
             role: 'user',
             content: `What is the capital of ${input['country']}?`
           }
         ]
       });
       return response.choices[0].message.content;
     },
  metrics: ['accuracy'],
  projectName: 'my-project'
});

// Run an experiment with a prompt template
const promptTemplate = await createPromptTemplate({
  template: [{ role: 'user', content: 'What is the capital of {{ country }}?' }],
  name: 'my-prompt-template',
  projectName: 'my-project'
});

const results = await runExperiment({
  name: 'my-experiment',
  dataset: [{ country: 'France' }],
  promptTemplate
  metrics: ['accuracy'],
  projectName: 'my-project'
});
```

The project can be specified by providing exactly one of the project name
(via the 'project' parameter or the GALILEO\_PROJECT environment variable)
or the project ID (via the 'project\_id' parameter or the GALILEO\_PROJECT\_ID environment variable).

## Type Parameters

### T

`T` *extends* `Record`\<`string`, `unknown`>

## Parameters

### params

`RunExperimentParams`\<`T`>

## Returns

`Promise`\<`RunExperimentOutput`>

Array of outputs from the processing function