math_dataset

References:

algebra__linear_1d

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/algebra__linear_1d')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

algebra__linear_1d_composed

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/algebra__linear_1d_composed')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

algebra__linear_2d

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/algebra__linear_2d')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

algebra__linear_2d_composed

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/algebra__linear_2d_composed')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

algebra__polynomial_roots

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/algebra__polynomial_roots')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

algebra__polynomial_roots_composed

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/algebra__polynomial_roots_composed')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

algebra__sequence_next_term

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/algebra__sequence_next_term')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

algebra__sequence_nth_term

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/algebra__sequence_nth_term')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

arithmetic__add_or_sub

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/arithmetic__add_or_sub')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

arithmetic__add_or_sub_in_base

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/arithmetic__add_or_sub_in_base')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

arithmetic__add_sub_multiple

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/arithmetic__add_sub_multiple')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

arithmetic__div

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/arithmetic__div')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

arithmetic__mixed

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/arithmetic__mixed')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

arithmetic__mul

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/arithmetic__mul')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

arithmetic__mul_div_multiple

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/arithmetic__mul_div_multiple')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

arithmetic__nearest_integer_root

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/arithmetic__nearest_integer_root')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

arithmetic__simplify_surd

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/arithmetic__simplify_surd')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

calculus__differentiate

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/calculus__differentiate')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

calculus__differentiate_composed

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/calculus__differentiate_composed')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

comparison__closest

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/comparison__closest')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

comparison__closest_composed

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/comparison__closest_composed')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

comparison__kth_biggest

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/comparison__kth_biggest')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

comparison__kth_biggest_composed

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/comparison__kth_biggest_composed')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

comparison__pair

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/comparison__pair')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

comparison__pair_composed

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/comparison__pair_composed')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

comparison__sort

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/comparison__sort')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

comparison__sort_composed

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/comparison__sort_composed')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

measurement__conversion

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/measurement__conversion')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

measurement__time

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/measurement__time')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

numbers__base_conversion

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__base_conversion')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

numbers__div_remainder

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__div_remainder')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

numbers__div_remainder_composed

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__div_remainder_composed')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

numbers__gcd

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__gcd')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

numbers__gcd_composed

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__gcd_composed')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

numbers__is_factor

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__is_factor')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

numbers__is_factor_composed

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__is_factor_composed')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

numbers__is_prime

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__is_prime')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

numbers__is_prime_composed

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__is_prime_composed')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

numbers__lcm

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__lcm')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

numbers__lcm_composed

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__lcm_composed')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

numbers__list_prime_factors

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__list_prime_factors')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

numbers__list_prime_factors_composed

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__list_prime_factors_composed')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

numbers__place_value

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__place_value')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

numbers__place_value_composed

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__place_value_composed')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

numbers__round_number

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__round_number')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

numbers__round_number_composed

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__round_number_composed')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

polynomials__add

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/polynomials__add')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

polynomials__coefficient_named

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/polynomials__coefficient_named')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

polynomials__collect

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/polynomials__collect')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

polynomials__compose

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/polynomials__compose')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

polynomials__evaluate

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/polynomials__evaluate')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

polynomials__evaluate_composed

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/polynomials__evaluate_composed')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

polynomials__expand

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/polynomials__expand')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

polynomials__simplify_power

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/polynomials__simplify_power')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

probability__swr_p_level_set

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/probability__swr_p_level_set')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

probability__swr_p_sequence

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:math_dataset/probability__swr_p_sequence')
  • Description:
Mathematics database.

This dataset code generates mathematical question and answer pairs,
from a range of question types at roughly school-level difficulty.
This is designed to test the mathematical learning and algebraic
reasoning skills of learning models.

Original paper: Analysing Mathematical Reasoning Abilities of Neural Models
(Saxton, Grefenstette, Hill, Kohli).

Example usage:
train_examples, val_examples = datasets.load_dataset(
    'math_dataset/arithmetic__mul',
    split=['train', 'test'],
    as_supervised=True)
  • License: No known license
  • Version: 1.0.0
  • Splits:
Split Examples
'test' 10000
'train' 1999998
  • Features:
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}