scan

References:

simple

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:scan/simple')
  • Description:
SCAN tasks with various splits.

SCAN is a set of simple language-driven navigation tasks for studying
compositional learning and zero-shot generalization.

See https://github.com/brendenlake/SCAN for a description of the splits.

Example usage:
data = datasets.load_dataset('scan/length')
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 4182
'train' 16728
  • Features:
{
    "commands": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "actions": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

addprim_jump

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:scan/addprim_jump')
  • Description:
SCAN tasks with various splits.

SCAN is a set of simple language-driven navigation tasks for studying
compositional learning and zero-shot generalization.

See https://github.com/brendenlake/SCAN for a description of the splits.

Example usage:
data = datasets.load_dataset('scan/length')
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 7706
'train' 14670
  • Features:
{
    "commands": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "actions": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

addprim_turn_left

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:scan/addprim_turn_left')
  • Description:
SCAN tasks with various splits.

SCAN is a set of simple language-driven navigation tasks for studying
compositional learning and zero-shot generalization.

See https://github.com/brendenlake/SCAN for a description of the splits.

Example usage:
data = datasets.load_dataset('scan/length')
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 1208
'train' 21890
  • Features:
{
    "commands": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "actions": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

filler_num0

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:scan/filler_num0')
  • Description:
SCAN tasks with various splits.

SCAN is a set of simple language-driven navigation tasks for studying
compositional learning and zero-shot generalization.

See https://github.com/brendenlake/SCAN for a description of the splits.

Example usage:
data = datasets.load_dataset('scan/length')
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 1173
'train' 15225
  • Features:
{
    "commands": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "actions": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

filler_num1

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:scan/filler_num1')
  • Description:
SCAN tasks with various splits.

SCAN is a set of simple language-driven navigation tasks for studying
compositional learning and zero-shot generalization.

See https://github.com/brendenlake/SCAN for a description of the splits.

Example usage:
data = datasets.load_dataset('scan/length')
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 1173
'train' 16290
  • Features:
{
    "commands": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "actions": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

filler_num2

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:scan/filler_num2')
  • Description:
SCAN tasks with various splits.

SCAN is a set of simple language-driven navigation tasks for studying
compositional learning and zero-shot generalization.

See https://github.com/brendenlake/SCAN for a description of the splits.

Example usage:
data = datasets.load_dataset('scan/length')
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 1173
'train' 17391
  • Features:
{
    "commands": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "actions": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

filler_num3

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:scan/filler_num3')
  • Description:
SCAN tasks with various splits.

SCAN is a set of simple language-driven navigation tasks for studying
compositional learning and zero-shot generalization.

See https://github.com/brendenlake/SCAN for a description of the splits.

Example usage:
data = datasets.load_dataset('scan/length')
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 1173
'train' 18528
  • Features:
{
    "commands": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "actions": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

length

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:scan/length')
  • Description:
SCAN tasks with various splits.

SCAN is a set of simple language-driven navigation tasks for studying
compositional learning and zero-shot generalization.

See https://github.com/brendenlake/SCAN for a description of the splits.

Example usage:
data = datasets.load_dataset('scan/length')
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 3920
'train' 16990
  • Features:
{
    "commands": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "actions": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

template_around_right

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:scan/template_around_right')
  • Description:
SCAN tasks with various splits.

SCAN is a set of simple language-driven navigation tasks for studying
compositional learning and zero-shot generalization.

See https://github.com/brendenlake/SCAN for a description of the splits.

Example usage:
data = datasets.load_dataset('scan/length')
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 4476
'train' 15225
  • Features:
{
    "commands": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "actions": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

template_jump_around_right

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:scan/template_jump_around_right')
  • Description:
SCAN tasks with various splits.

SCAN is a set of simple language-driven navigation tasks for studying
compositional learning and zero-shot generalization.

See https://github.com/brendenlake/SCAN for a description of the splits.

Example usage:
data = datasets.load_dataset('scan/length')
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 1173
'train' 18528
  • Features:
{
    "commands": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "actions": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

template_opposite_right

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:scan/template_opposite_right')
  • Description:
SCAN tasks with various splits.

SCAN is a set of simple language-driven navigation tasks for studying
compositional learning and zero-shot generalization.

See https://github.com/brendenlake/SCAN for a description of the splits.

Example usage:
data = datasets.load_dataset('scan/length')
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 4476
'train' 15225
  • Features:
{
    "commands": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "actions": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

template_right

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:scan/template_right')
  • Description:
SCAN tasks with various splits.

SCAN is a set of simple language-driven navigation tasks for studying
compositional learning and zero-shot generalization.

See https://github.com/brendenlake/SCAN for a description of the splits.

Example usage:
data = datasets.load_dataset('scan/length')
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 4476
'train' 15225
  • Features:
{
    "commands": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "actions": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}