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

# Dataset

> Use FutureAGI Dataset to create and manage your datasets

You can checkout the [colab notebook](https://colab.research.google.com/drive/1TCRKBGoVOmdjNm60HHH1LeGGBbWAvN2L?usp=sharing) to quickly get started with the FutureAGI Dataset. <a href="https://colab.research.google.com/drive/1TCRKBGoVOmdjNm60HHH1LeGGBbWAvN2L?usp=sharing"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab" /></a>

## Installing FutureAGI SDK

```bash theme={null}
pip install futureagi
```

## Initializing FutureAGI Dataset

```python theme={null}
from fi.datasets import Dataset

dataset = Dataset(fi_api_key="<your_api_key>", 
                  fi_secret_key="<your_api_secret>") # Optional, if you want to set the API key and secret key manually
```

<Tip>
  Click [here](/admin-settings#accessing-api-keys) to learn how to access your API keys.
  It's recommended to set the API key and secret key as environment variables.
</Tip>

## Create a Dataset

```python theme={null}
from fi.datasets import Dataset, DatasetConfig, ModelTypes
from fi.datasets.models import Column, Row, Cell, DataTypeChoices, SourceChoices
import uuid

# Create a dataset configuration
config = DatasetConfig(
    id=None,  # Will be set by the server
    name="my_dataset",  # Choose a unique name
    model_type=ModelTypes.GENERATIVE_LLM
)

# Initialize and create the dataset
dataset = Dataset(dataset_config=config)
dataset = dataset.create()
```

## Add Columns to Dataset

```python theme={null}
# Define columns
columns = [
    Column(
        name="Name",
        data_type=DataTypeChoices.TEXT,
        source=SourceChoices.OTHERS,
        source_id=None,
    ),
    Column(
        name="Age",
        data_type=DataTypeChoices.INTEGER,
        source=SourceChoices.OTHERS,
        source_id=None,
    ),
    Column(
        name="AUDIO_URLS",
        data_type=DataTypeChoices.AUDIO,
        source=SourceChoices.OTHERS,
        source_id=None
    )
]

# Add columns to dataset
dataset = dataset.add_columns(columns=columns)
```

## Add Rows to Dataset

```python theme={null}
# Define rows with cells
rows = [
    Row(
        order=1,
        cells=[
            Cell(column_name="Name", value="Alice"),
            Cell(column_name="Age", value=25),
            Cell(column_name="AUDIO_URLS", value="https://example.com/audio1.mp3")
        ],
    ),
    Row(
        order=2,
        cells=[
            Cell(column_name="Name", value="Bob"),
            Cell(column_name="Age", value=30),
            Cell(column_name="AUDIO_URLS", value="https://example.com/audio2.mp3")
        ],
    ),
]

# Add rows to dataset
dataset = dataset.add_rows(rows=rows)
```

## Download Dataset

```python theme={null}
# Download dataset to a CSV file
file_path = "my_dataset.csv"
dataset.download(file_path=file_path)

# Read the downloaded file
with open(file_path, "r") as file:
    content = file.read()
    print(content)
```

## Delete Dataset

```python theme={null}
# Delete the dataset
dataset.delete()
```

<Tip>
  Make sure to handle the downloaded file cleanup after you're done with it:

  ```python theme={null}
  import os

  if os.path.exists(file_path):
      os.remove(file_path)
  ```
</Tip>