math_dataset

مراجع:

الجبر__الخطي_1د

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/algebra__linear_1d')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

الجبر__الخطي_1d_المركب

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/algebra__linear_1d_composed')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

الجبر__الخطي_2د

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/algebra__linear_2d')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

الجبر__الخطي_2d_المركب

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/algebra__linear_2d_composed')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

الجبر__جذور_متعددة_الحدود

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/algebra__polynomial_roots')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

الجبر__متعددة_الحدود_جذور_مركبة

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/algebra__polynomial_roots_composed')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

الجبر__التسلسل_التالي_المصطلح

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/algebra__sequence_next_term')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

الجبر__التسلسل_nth_term

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/algebra__sequence_nth_term')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

الحسابية__add_or_sub

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/arithmetic__add_or_sub')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

الحسابية__add_or_sub_in_base

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/arithmetic__add_or_sub_in_base')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

الحسابي__add_sub_multiple

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/arithmetic__add_sub_multiple')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

الحساب__div

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/arithmetic__div')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

الحساب__مختلط

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/arithmetic__mixed')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

الحساب__mul

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/arithmetic__mul')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

arithmetic__mul_div_multiple

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/arithmetic__mul_div_multiple')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

arithmetic__nearest_integer_root

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/arithmetic__nearest_integer_root')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

arithmetic__simplify_surd

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/arithmetic__simplify_surd')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

حساب التفاضل والتكامل__التمايز

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/calculus__differentiate')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

حساب التفاضل والتكامل__ التفاضل_المركب

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/calculus__differentiate_composed')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

مقارنة__الأقرب

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/comparison__closest')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

مقارنة__أقرب_مركب

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/comparison__closest_composed')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

مقارنة__كث_أكبر

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/comparison__kth_biggest')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

مقارنة__كث_أكبر_مركب

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/comparison__kth_biggest_composed')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

مقارنة__زوج

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/comparison__pair')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

match__pair_compose

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/comparison__pair_composed')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

مقارنة__فرز

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/comparison__sort')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

بالمقارنة__الفرز_المركب

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/comparison__sort_composed')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

قياس__تحويل

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/measurement__conversion')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

قياس__الوقت

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/measurement__time')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

number__base_conversion

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__base_conversion')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

number__div_remainder

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__div_remainder')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

number__div_remainder_compose

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__div_remainder_composed')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

number__gcd

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__gcd')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

number__gcd_compose

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__gcd_composed')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

number__is_factor

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__is_factor')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

number__is_factor_compose

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__is_factor_composed')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

number__is_prime

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__is_prime')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

number__is_prime_compose

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__is_prime_composed')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

number__lcm

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__lcm')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

number__lcm_compose

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__lcm_composed')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

number__list_prime_factors

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__list_prime_factors')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

Numbers__list_prime_factors_compose

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__list_prime_factors_composed')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

number__place_value

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__place_value')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

Numbers__place_value_compose

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__place_value_composed')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

number__round_number

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__round_number')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

number__round_number_compose

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/numbers__round_number_composed')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

كثيرات الحدود__add

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/polynomials__add')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

كثيرات الحدود__معامل_المسمى

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/polynomials__coefficient_named')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

كثيرات الحدود__collect

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/polynomials__collect')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

كثيرات الحدود__compose

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/polynomials__compose')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

كثيرات الحدود__تقييم

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/polynomials__evaluate')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

كثيرات الحدود__تقييم_المركبة

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/polynomials__evaluate_composed')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

كثيرات الحدود__توسيع

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/polynomials__expand')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

كثيرات الحدود__تبسيط_القوة

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/polynomials__simplify_power')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

احتمال__swr_p_level_set

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/probability__swr_p_level_set')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

احتمال__swr_p_sequence

استخدم الأمر التالي لتحميل مجموعة البيانات هذه في TFDS:

ds = tfds.load('huggingface:math_dataset/probability__swr_p_sequence')
  • وصف :
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)
  • الترخيص : لا يوجد ترخيص معروف
  • الإصدار : 1.0.0
  • الإنشقاقات :
ينقسم أمثلة
'test' 10000
'train' 1999998
  • سمات :
{
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answer": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}