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 = tfds.load(
'math*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

- URL: https://github.com/deepmind/mathematics_dataset
`DatasetBuilder`

:`tfds.text.math_dataset.MathDataset`

`math_dataset`

is configured with `tfds.core.dataset_builder.BuilderConfig`

and
has the following configurations predefined (defaults to the first one):

`algebra__linear_1d`

(`v1.0.0`

) (`Size: 2.17 GiB`

): 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 = tfds.load(
'math*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`algebra__linear_1d_composed`

(`v1.0.0`

) (`Size: 2.17 GiB`

): 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 = tfds.load(
'math*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`algebra__linear_2d`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`algebra__linear_2d_composed`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`algebra__polynomial_roots`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`algebra__polynomial_roots_composed`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`algebra__sequence_next_term`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`algebra__sequence_nth_term`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`arithmetic__add_or_sub`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`arithmetic__add_or_sub_in_base`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`arithmetic__add_sub_multiple`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`arithmetic__div`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`arithmetic__mixed`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`arithmetic__mul`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`arithmetic__mul_div_multiple`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`arithmetic__nearest_integer_root`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`arithmetic__simplify_surd`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`calculus__differentiate`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`calculus__differentiate_composed`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`comparison__closest`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`comparison__closest_composed`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`comparison__kth_biggest`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`comparison__kth_biggest_composed`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`comparison__pair`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`comparison__pair_composed`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`comparison__sort`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`comparison__sort_composed`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`measurement__conversion`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`measurement__time`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`numbers__base_conversion`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`numbers__div_remainder`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`numbers__div_remainder_composed`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`numbers__gcd`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`numbers__gcd_composed`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`numbers__is_factor`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`numbers__is_factor_composed`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`numbers__is_prime`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`numbers__is_prime_composed`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`numbers__lcm`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`numbers__lcm_composed`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`numbers__list_prime_factors`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`numbers__list_prime_factors_composed`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`numbers__place_value`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`numbers__place_value_composed`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`numbers__round_number`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`numbers__round_number_composed`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`polynomials__add`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`polynomials__coefficient_named`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`polynomials__collect`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`polynomials__compose`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`polynomials__evaluate`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`polynomials__evaluate_composed`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`polynomials__expand`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`polynomials__simplify_power`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`probability__swr_p_level_set`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`probability__swr_p_sequence`

(`v1.0.0`

) (`Size: 2.17 GiB`

): Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

`math_dataset/algebra__linear_1d`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/algebra__linear_1d_composed`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/algebra__linear_2d`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/algebra__linear_2d_composed`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/algebra__polynomial_roots`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/algebra__polynomial_roots_composed`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/algebra__sequence_next_term`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/algebra__sequence_nth_term`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/arithmetic__add_or_sub`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/arithmetic__add_or_sub_in_base`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/arithmetic__add_sub_multiple`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/arithmetic__div`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/arithmetic__mixed`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/arithmetic__mul`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/arithmetic__mul_div_multiple`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/arithmetic__nearest_integer_root`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/arithmetic__simplify_surd`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/calculus__differentiate`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/calculus__differentiate_composed`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/comparison__closest`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/comparison__closest_composed`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/comparison__kth_biggest`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/comparison__kth_biggest_composed`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/comparison__pair`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/comparison__pair_composed`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/comparison__sort`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/comparison__sort_composed`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/measurement__conversion`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/measurement__time`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/numbers__base_conversion`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/numbers__div_remainder`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/numbers__div_remainder_composed`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/numbers__gcd`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/numbers__gcd_composed`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/numbers__is_factor`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/numbers__is_factor_composed`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/numbers__is_prime`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/numbers__is_prime_composed`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/numbers__lcm`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/numbers__lcm_composed`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/numbers__list_prime_factors`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/numbers__list_prime_factors_composed`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/numbers__place_value`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/numbers__place_value_composed`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/numbers__round_number`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/numbers__round_number_composed`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/polynomials__add`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/polynomials__coefficient_named`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/polynomials__collect`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/polynomials__compose`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/polynomials__evaluate`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/polynomials__evaluate_composed`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/polynomials__expand`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/polynomials__simplify_power`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/probability__swr_p_level_set`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

`math_dataset/probability__swr_p_sequence`

Mathematics database.

*dataset/arithmetic*_mul', split=['train', 'test'], as_supervised=True)

Versions:

(default):`1.0.0`

### Statistics

Split | Examples |
---|---|

ALL | 2,009,998 |

TRAIN | 1,999,998 |

TEST | 10,000 |

### Features

```
FeaturesDict({
'answer': Text(shape=(), dtype=tf.string),
'question': Text(shape=(), dtype=tf.string),
})
```

### Homepage

### Supervised keys (for `as_supervised=True`

)

`('question', 'answer')`

## Citation

```
@article{2019arXiv,
author = {Saxton, Grefenstette, Hill, Kohli},
title = {Analysing Mathematical Reasoning Abilities of Neural Models},
year = {2019},
journal = {arXiv:1904.01557}
}
```