pkg_pyknnclassifier.predict

Module Contents

Functions

predict(train_X, train_y, unlabel_df[, pred_method, k])

This function predicts the labels of the unlabled observations based on the similarity score calculated from Euclidean distance.
pkg_pyknnclassifier.predict.predict(train_X, train_y, unlabel_df, pred_method='hard', k=3)[source]

This function predicts the labels of the unlabled observations based on the similarity score calculated from Euclidean distance.

Parameters:

  • train_X (pd.DataFrame) – The data frame containing labeled observations, but without the label.

  • train_y (numpy.array) – The array containing labels in the training dataset

  • unlabel_df (pd.DataFrame) – The data frame containing unlabeld observations.

  • pred_method (str) – ‘soft’ or ‘hard’.

  • k (int) – The number of nearest neighbors to consider for making predictions.

Returns:

An array containing predicted labels for the observations.

Return type:

array

Examples

df = pd.DataFrame({‘A’:[0.5, 0.2, 0.4],

‘B’:[0.3, 0.2, 0.5]})

predict(df)