Data Binning

Endpoint: POST /binning

Request Parameters

File Upload

• file (required): The CSV file to be processed.

Form Parameters

• column (required): The column to apply binning.
• method (optional, default="equal_width"): Binning method to apply. Supported values:
  • "equal_width": Divides data into bins of equal width.
  • "equal_frequency": Divides data into bins with an equal number of data points.
  • "custom": Uses custom-defined bin edges.
• num_bins (optional, default=5): Number of bins for equal-width or equal-frequency methods.
• custom_bins (optional): Comma-separated list of custom bin edges.
• labels (optional): Comma-separated list of bin labels.
• include_lowest (optional, default=True): Whether to include the lowest value in the first bin.
• right (optional, default=True): Whether the bins are right-inclusive.
• handle_outliers (optional, default="other"): How to handle outliers. Options:
  • "other": Assigns outliers to separate bins.
  • "exclude": Excludes outliers from binning.
• dtype (optional, default="categorical"): Output data type of the binned column. Supported values: "categorical", "integer", "string".

Processing Logic

1. File Reading: The uploaded file is read and converted into a Pandas DataFrame.
2. Input Validation: Ensures the specified column exists and contains numeric data.
3. Method Selection:
   • Equal Width: Creates equal-width bins based on data range.
   • Equal Frequency: Distributes data into bins with equal frequencies.
   • Custom Binning: Applies user-defined bin edges.
4. Handle Outliers: Configures the handling of outliers based on user preference.
5. Labeling and Data Type Conversion: Applies labels and converts the output to the specified data type.
6. Return Transformed File: Provides a downloadable file with the binned column appended.

Example Request

curl -X POST "<endpoint-url>/binning" \
  -F "file=@data.csv" \
  -F "column=values" \
  -F "method=equal_width" \
  -F "num_bins=4"