---
dataset_info:
  features:
  - name: age
    dtype: int64
  - name: job
    dtype:
      class_label:
        names:
          '0': admin.
          '1': blue-collar
          '2': entrepreneur
          '3': housemaid
          '4': management
          '5': retired
          '6': self-employed
          '7': services
          '8': student
          '9': technician
          '10': unemployed
          '11': unknown
  - name: marital
    dtype:
      class_label:
        names:
          '0': divorced
          '1': married
          '2': single
          '3': unknown
  - name: education
    dtype:
      class_label:
        names:
          '0': primary
          '1': secondary
          '2': tertiary
          '3': unknown
          '4': basic.4y
          '5': basic.6y
          '6': basic.9y
          '7': high.school
          '8': illiterate
          '9': professional.course
          '10': university.degree
  - name: default
    dtype:
      class_label:
        names:
          '0': 'no'
          '1': 'yes'
          '2': unknown
  - name: housing
    dtype:
      class_label:
        names:
          '0': 'no'
          '1': 'yes'
          '2': unknown
  - name: loan
    dtype:
      class_label:
        names:
          '0': 'no'
          '1': 'yes'
          '2': unknown
  - name: contact
    dtype:
      class_label:
        names:
          '0': cellular
          '1': telephone
          '2': unknown
  - name: month
    dtype:
      class_label:
        names:
          '0': jan
          '1': feb
          '2': mar
          '3': apr
          '4': may
          '5': jun
          '6': jul
          '7': aug
          '8': sep
          '9': oct
          '10': nov
          '11': dec
  - name: day_of_week
    dtype:
      class_label:
        names:
          '0': mon
          '1': tue
          '2': wed
          '3': thu
          '4': fri
  - name: duration
    dtype: int64
  - name: campaign
    dtype: int64
  - name: pdays
    dtype: int64
  - name: previous
    dtype: int64
  - name: poutcome
    dtype:
      class_label:
        names:
          '0': failure
          '1': other
          '2': success
          '3': unknown
          '4': nonexistent
  - name: emp.var.rate
    dtype: float32
  - name: cons.price.idx
    dtype: float32
  - name: cons.conf.idx
    dtype: float32
  - name: euribor3m
    dtype: float32
  - name: nr.employed
    dtype: float32
  - name: y
    dtype:
      class_label:
        names:
          '0': 'no'
          '1': 'yes'
  splits:
  - name: train
    num_bytes: 6095824
    num_examples: 41188
  download_size: 469576
  dataset_size: 6095824
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---

# Dataset Card for Bank Marketing (additional)

This dataset is a precise version of [UCI Bank Marketing](https://archive.ics.uci.edu/dataset/222/bank+marketing)

We first created the default bank marketing dataset, as seen [here](https://huggingface.co/datasets/cestwc/bank-marketing). Then we further run the following Python script to create this additional portion.

```python
# Define feature types
continuous_columns = ["age", "duration", "campaign", "pdays", "previous",
                      "emp.var.rate", "cons.price.idx", "cons.conf.idx",
                      "euribor3m", "nr.employed"]

categorical_columns = ["job", "marital", "education", "default", "housing", "loan",
                       "contact", "month", "day_of_week", "poutcome", "y"]

# Extract category mappings from the reference dataset (bank-additional)
category_mappings_additional = {col: reference_categories[col] for col in categorical_columns}


hf_features_additional = Features({
    "age": Value("int64"),
    "job": ClassLabel(names=category_mappings_additional["job"]),
    "marital": ClassLabel(names=category_mappings_additional["marital"]),
    "education": ClassLabel(names=category_mappings_additional["education"]),
    "default": ClassLabel(names=category_mappings_additional["default"]),
    "housing": ClassLabel(names=category_mappings_additional["housing"]),
    "loan": ClassLabel(names=category_mappings_additional["loan"]),
    "contact": ClassLabel(names=category_mappings_additional["contact"]),
    "month": ClassLabel(names=category_mappings_additional["month"]),
    "day_of_week": ClassLabel(names=category_mappings_additional["day_of_week"]),
    "duration": Value("int64"),
    "campaign": Value("int64"),
    "pdays": Value("int64"),
    "previous": Value("int64"),
    "poutcome": ClassLabel(names=category_mappings_additional["poutcome"]),
    "emp.var.rate": Value("float32"),
    "cons.price.idx": Value("float32"),
    "cons.conf.idx": Value("float32"),
    "euribor3m": Value("float32"),
    "nr.employed": Value("float32"),
    "y": ClassLabel(names=category_mappings_additional["y"])  # Target column
})

# Convert pandas DataFrame to Hugging Face Dataset
hf_dataset_additional = Dataset.from_pandas(df_additional, features=hf_features_additional)

# Print dataset structure
print(hf_dataset_additional)
```

The printed output could look like

```
Dataset({
    features: ['age', 'job', 'marital', 'education', 'default', 'housing', 'loan', 'contact', 'month', 'day_of_week', 'duration', 'campaign', 'pdays', 'previous', 'poutcome', 'emp.var.rate', 'cons.price.idx', 'cons.conf.idx', 'euribor3m', 'nr.employed', 'y'],
    num_rows: 41188
})
```