How to use the pdpipe.AggByCols function in pdpipe

To help you get started, weโ€™ve selected a few pdpipe examples, based on popular ways it is used in public projects.

Secure your code as it's written. Use Snyk Code to scan source code in minutes - no build needed - and fix issues immediately.

github pdpipe / pdpipe / tests / col_generation / test_aggbycols.py View on Github external
def test_aggbycols_with_result_columns():
    """Testing AggByCols pipeline stages."""
    df = ph_df()
    round_ph = AggByCols("ph", np.log, result_columns='log_ph')
    res_df = round_ph(df)
    assert 'ph' not in res_df.columns
    assert res_df.columns.get_loc('log_ph') == 0
    assert_approx_equal(res_df['log_ph'][1], _LOG32, significant=5)
    assert_approx_equal(res_df['log_ph'][2], _LOG72, significant=5)
    assert_approx_equal(res_df['log_ph'][3], _LOG121, significant=5)
github pdpipe / pdpipe / tests / col_generation / test_aggbycols.py View on Github external
def test_aggbycols_no_drop_custom_suffix():
    """Testing AggByCols pipeline stages."""
    df = ph_df()
    round_ph = AggByCols("ph", np.log, drop=False, suffix='_log')
    res_df = round_ph(df)
    assert 'ph' in res_df.columns
    assert 'ph_log' in res_df.columns
    assert res_df.columns.get_loc('ph') == 0
    assert res_df.columns.get_loc('ph_log') == 1
    assert_approx_equal(res_df['ph_log'][1], _LOG32, significant=5)
    assert_approx_equal(res_df['ph_log'][2], _LOG72, significant=5)
    assert_approx_equal(res_df['ph_log'][3], _LOG121, significant=5)
github pdpipe / pdpipe / tests / col_generation / test_aggbycols.py View on Github external
def test_aggbycols_with_drop():
    """Testing AggByCols pipeline stages."""
    df = ph_df()
    round_ph = AggByCols("ph", np.log, drop=False)
    res_df = round_ph(df)
    assert 'ph' in res_df.columns
    assert 'ph_agg' in res_df.columns
    assert res_df.columns.get_loc('ph') == 0
    assert res_df.columns.get_loc('ph_agg') == 1
    assert_approx_equal(res_df['ph_agg'][1], _LOG32, significant=5)
    assert_approx_equal(res_df['ph_agg'][2], _LOG72, significant=5)
    assert_approx_equal(res_df['ph_agg'][3], _LOG121, significant=5)
github pdpipe / pdpipe / tests / col_generation / test_aggbycols.py View on Github external
def test_aggbycols_func_desc():
    """Testing AggByCols pipeline stages."""
    df = ph_df()
    round_ph = AggByCols("ph", np.log, func_desc='Round PH values')
    res_df = round_ph(df)
    assert res_df.columns.get_loc('ph') == 0
    assert_approx_equal(res_df['ph'][1], _LOG32, significant=5)
    assert_approx_equal(res_df['ph'][2], _LOG72, significant=5)
    assert_approx_equal(res_df['ph'][3], _LOG121, significant=5)
github pdpipe / pdpipe / tests / col_generation / test_aggbycols.py View on Github external
def test_aggbycols():
    """Testing AggByCols pipeline stages."""
    df = ph_df()
    round_ph = AggByCols("ph", np.log)
    res_df = round_ph(df)
    assert res_df.columns.get_loc('ph') == 0
    assert_approx_equal(res_df['ph'][1], _LOG32, significant=5)
    assert_approx_equal(res_df['ph'][2], _LOG72, significant=5)
    assert_approx_equal(res_df['ph'][3], _LOG121, significant=5)
github pdpipe / pdpipe / tests / col_generation / test_aggbycols.py View on Github external
def test_aggbycols_with_bad_len_result_columns():
    """Testing ApplyByCols pipeline stages."""
    with pytest.raises(ValueError):
        AggByCols("ph", np.log, result_columns=['a', 'b'])