Encoding Categorical Data

Endpoint: POST /encoding

Request Parameters

File Upload

• file (required): The file to be processed. Supported formats include CSV, Excel, etc.

Form Parameters

• method (required): The encoding method to apply. Supported values:
  • onehot: One-hot encoding.
  • label: Label encoding.
  • target: Target encoding (requires a target column).
  • ordinal: Ordinal encoding.
  • frequency: Frequency encoding.
• target_column (optional): The target column required for target encoding. This parameter is mandatory if method is target.

Processing Logic

1. Read the uploaded file: The file is converted into a Pandas DataFrame.
2. Identify categorical columns: Selects columns with the object data type (categorical columns).
3. Validate the encoding method: Ensures the specified method is supported.
4. Apply the encoding:
   • One-Hot Encoding: Converts categorical columns into binary columns (one-hot encoding).
   • Label Encoding: Converts categorical values into integer labels.
   • Target Encoding: Replaces categorical values with the mean of the target column for each category.
   • Ordinal Encoding: Converts categorical values into ordinal integers.
   • Frequency Encoding: Replaces categorical values with their frequency in the dataset.
5. Return the modified file: The encoded file is provided for download with the filename appended with _encoded_file.

Example Request

curl -X POST "http://localhost:8000/encoding" \
-F "file=@data.csv" \
-F "method=onehot"