german-credit / README.md
cestwc's picture
Update README.md
c942395 verified
metadata
dataset_info:
  features:
    - name: status of existing checking account
      dtype:
        class_label:
          names:
            '0': < 0 DM
            '1': 0 <= ... < 200 DM
            '2': '>= 200 DM / salary assignments for at least 1 year'
            '3': no checking account
    - name: duration in month
      dtype: int64
    - name: credit history
      dtype:
        class_label:
          names:
            '0': no credits taken / all credits paid back duly
            '1': all credits at this bank paid back duly
            '2': existing credits paid back duly till now
            '3': delay in paying off in the past
            '4': critical account / other credits existing (not at this bank)
    - name: purpose
      dtype:
        class_label:
          names:
            '0': car (new)
            '1': car (used)
            '2': furniture/equipment
            '3': radio/television
            '4': domestic appliances
            '5': repairs
            '6': education
            '7': vacation
            '8': retraining
            '9': business
            '10': others
    - name: credit amount
      dtype: int64
    - name: savings account/bonds
      dtype:
        class_label:
          names:
            '0': < 100 DM
            '1': 100 <= ... < 500 DM
            '2': 500 <= ... < 1000 DM
            '3': '>= 1000 DM'
            '4': unknown / no savings account
    - name: present employment since
      dtype:
        class_label:
          names:
            '0': unemployed
            '1': < 1 year
            '2': 1 <= ... < 4 years
            '3': 4 <= ... < 7 years
            '4': '>= 7 years'
    - name: installment rate in percentage of disposable income
      dtype: int64
    - name: personal status and sex
      dtype:
        class_label:
          names:
            '0': 'male: divorced/separated'
            '1': 'female: divorced/separated/married'
            '2': 'male: single'
            '3': 'male: married/widowed'
            '4': 'female: single'
    - name: other debtors / guarantors
      dtype:
        class_label:
          names:
            '0': none
            '1': co-applicant
            '2': guarantor
    - name: present residence since
      dtype: int64
    - name: property
      dtype:
        class_label:
          names:
            '0': real estate
            '1': building society savings agreement / life insurance
            '2': car or other, not in attribute 6
            '3': unknown / no property
    - name: age in years
      dtype: int64
    - name: other installment plans
      dtype:
        class_label:
          names:
            '0': bank
            '1': stores
            '2': none
    - name: housing
      dtype:
        class_label:
          names:
            '0': rent
            '1': own
            '2': for free
    - name: number of existing credits at this bank
      dtype: int64
    - name: job
      dtype:
        class_label:
          names:
            '0': unemployed / unskilled - non-resident
            '1': unskilled - resident
            '2': skilled employee / official
            '3': management / self-employed / highly qualified employee / officer
    - name: number of people being liable to provide maintenance for
      dtype: int64
    - name: telephone
      dtype:
        class_label:
          names:
            '0': none
            '1': yes, registered under the customer’s name
    - name: foreign worker
      dtype:
        class_label:
          names:
            '0': 'yes'
            '1': 'no'
    - name: class
      dtype:
        class_label:
          names:
            '0': good
            '1': bad
  splits:
    - name: train
      num_bytes: 168000
      num_examples: 1000
  download_size: 25200
  dataset_size: 168000
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

Dataset Card for Dataset Name

This dataset is a precise version of Statlog (German Credit Data), donated on 11/16/1994.

We used the following Python script to produce this Hugging Face dataset.

columns = [
    "status of existing checking account",  # 1 (categorical)
    "duration in month",  # 2 (numerical)
    "credit history",  # 3 (categorical)
    "purpose",  # 4 (categorical)
    "credit amount",  # 5 (numerical)
    "savings account/bonds",  # 6 (categorical)
    "present employment since",  # 7 (categorical)
    "installment rate in percentage of disposable income",  # 8 (numerical)
    "personal status and sex",  # 9 (categorical)
    "other debtors / guarantors",  # 10 (categorical)
    "present residence since",  # 11 (numerical)
    "property",  # 12 (categorical)
    "age in years",  # 13 (numerical)
    "other installment plans",  # 14 (categorical)
    "housing",  # 15 (categorical)
    "number of existing credits at this bank",  # 16 (numerical)
    "job",  # 17 (categorical)
    "number of people being liable to provide maintenance for",  # 18 (numerical)
    "telephone",  # 19 (categorical)
    "foreign worker",  # 20 (categorical)
    "class"  # 21 (target label, categorical)
]


continuous_columns = [
    "duration in month", "credit amount",
    "installment rate in percentage of disposable income",
    "present residence since", "age in years",
    "number of existing credits at this bank",
    "number of people being liable to provide maintenance for"
]

categorical_columns = [
    "status of existing checking account", "credit history", "purpose",
    "savings account/bonds", "present employment since",
    "personal status and sex", "other debtors / guarantors", "property",
    "other installment plans", "housing", "job", "telephone",
    "foreign worker", "class"
]


import pandas as pd

# File path
file_path_categorical = "https://archive.ics.uci.edu/ml/machine-learning-databases/statlog/german/german.data"

# Load dataset
df_categorical = pd.read_csv(file_path_categorical, sep=" ", names=columns, skipinitialspace=True)

# Convert categorical features to integer indices
category_mappings = {}

for col in categorical_columns:
    df_categorical[col] = df_categorical[col].astype("category")
    category_mappings[col] = list(df_categorical[col].cat.categories)
    df_categorical[col] = df_categorical[col].cat.codes  # Convert to int indices

df_categorical = df_categorical[columns]


final_mappings = {
    "status of existing checking account": {
        "A11": "< 0 DM",
        "A12": "0 <= ... < 200 DM",
        "A13": ">= 200 DM / salary assignments for at least 1 year",
        "A14": "no checking account"
    },
    "credit history": {
        "A30": "no credits taken / all credits paid back duly",
        "A31": "all credits at this bank paid back duly",
        "A32": "existing credits paid back duly till now",
        "A33": "delay in paying off in the past",
        "A34": "critical account / other credits existing (not at this bank)"
    },
    "purpose": {
        "A40": "car (new)",
        "A41": "car (used)",
        "A42": "furniture/equipment",
        "A43": "radio/television",
        "A44": "domestic appliances",
        "A45": "repairs",
        "A46": "education",
        "A47": "vacation",
        "A48": "retraining",
        "A49": "business",
        "A410": "others"
    },
    "savings account/bonds": {
        "A61": "< 100 DM",
        "A62": "100 <= ... < 500 DM",
        "A63": "500 <= ... < 1000 DM",
        "A64": ">= 1000 DM",
        "A65": "unknown / no savings account"
    },
    "present employment since": {
        "A71": "unemployed",
        "A72": "< 1 year",
        "A73": "1 <= ... < 4 years",
        "A74": "4 <= ... < 7 years",
        "A75": ">= 7 years"
    },
    "personal status and sex": {
        "A91": "male: divorced/separated",
        "A92": "female: divorced/separated/married",
        "A93": "male: single",
        "A94": "male: married/widowed",
        "A95": "female: single"
    },
    "other debtors / guarantors": {
        "A101": "none",
        "A102": "co-applicant",
        "A103": "guarantor"
    },
    "property": {
        "A121": "real estate",
        "A122": "building society savings agreement / life insurance",
        "A123": "car or other, not in attribute 6",
        "A124": "unknown / no property"
    },
    "other installment plans": {
        "A141": "bank",
        "A142": "stores",
        "A143": "none"
    },
    "housing": {
        "A151": "rent",
        "A152": "own",
        "A153": "for free"
    },
    "job": {
        "A171": "unemployed / unskilled - non-resident",
        "A172": "unskilled - resident",
        "A173": "skilled employee / official",
        "A174": "management / self-employed / highly qualified employee / officer"
    },
    "telephone": {
        "A191": "none",
        "A192": "yes, registered under the customer’s name"
    },
    "foreign worker": {
        "A201": "yes",
        "A202": "no"
    },
    "class": {
        "1": "good",
        "2": "bad"
    }
}


from datasets import Dataset, DatasetDict, Features, Value, ClassLabel

hf_features_categorical = Features({
    col: Value("int64") if col in continuous_columns else
         ClassLabel(names=list(final_mappings[col].values())) if col in final_mappings else
         ClassLabel(names=["good", "bad"])  # "class" column
    for col in columns
})

from datasets import DatasetDict

# Convert pandas DataFrame to Hugging Face Dataset
df_categorical = df_categorical[columns]
hf_categorical = Dataset.from_pandas(df_categorical, features=hf_features_categorical)

# Store in a dataset dictionary
hf_dataset_categorical = DatasetDict({"train": hf_categorical})

# Print dataset structure
print(hf_dataset_categorical)

The printed output could look like

DatasetDict({
    train: Dataset({
        features: ['status of existing checking account', 'duration in month', 'credit history', 'purpose', 'credit amount', 'savings account/bonds', 'present employment since', 'installment rate in percentage of disposable income', 'personal status and sex', 'other debtors / guarantors', 'present residence since', 'property', 'age in years', 'other installment plans', 'housing', 'number of existing credits at this bank', 'job', 'number of people being liable to provide maintenance for', 'telephone', 'foreign worker', 'class'],
        num_rows: 1000
    })
})

Note that there is another file for numerical data, but seems that the content is not more than what we have so far, so we don't include it.