How to use register_dataframe_accessor method in pandera

Best Python code snippet using pandera_python

ext.py

Source:ext.py

1from typing import List, Tuple2import pandas as pd3@pd.api.extensions.register_dataframe_accessor("tag")4class CaTaggingAccessor:5    def __init__(self, df: pd.DataFrame):6        self._df = df7    def group_by_sentences(self):8        yield from (x[1] for x in self._df.groupby("sentence_id"))9    def group_by_documents(self):10        yield from (x[1] for x in self._df.groupby("document_id"))11    def number_of_sentences(self):12        return len(self._df.groupby("sentence_id"))13    def number_of_documents(self):14        return len(self._df.groupby("document_id"))15    def split_x_y_sentencewise(self) -> Tuple[List[List[str]], List[List[str]]]:16        X = []17        y = []18        for sent in self._df.tag.group_by_sentences():19            words = list(sent["word"])20            labels = list(sent["label"])21            X.append(words)22            y.append(labels)23        return X, y24    def get_times_per_document(self) -> List[int]:25        t = []26        # Right now, we assume that the time per token is the same27        # for a sentence. This might be an invalid assumption28        for df in self._df.tag.group_by_sentences():29            t.append(df["t"].values[0])30        return t31    def group_by_documents_x_y(self) -> Tuple[List[List[List[str]]], List[List[List[str]]]]:32        """Returns a list of documents that each contain a list of sentences and33        their respective labels grouped the same way.34        """35        X = []36        y = []37        for doc in self._df.tag.group_by_documents():38            X_doc = []39            y_doc = []40            for sent in doc.tag.group_by_sentences():41                words = list(sent["word"])42                labels = list(sent["label"])43                X_doc.append(words)44                y_doc.append(labels)45            X.append(X_doc)46            y.append(y_doc)47        return X, y48    def group_by_sentences_x_y(self) -> Tuple[List[List[str]], List[List[str]]]:49        """Returns a list of sentences and their respective labels grouped the same way."""50        X = []51        y = []52        for sent in self._df.tag.group_by_sentences():53            words = list(sent["word"])54            labels = list(sent["label"])55            assert len(words) == len(labels)56            X.append(words)57            y.append(labels)58        return X, y59@pd.api.extensions.register_dataframe_accessor("dclass")60class CaDocumentClassificationAccessor:61    def __init__(self, df: pd.DataFrame):62        self._df = df63    def split_x_y(self) -> Tuple[List[str], List[str]]:64        X = self._df["sentence"]65        y = self._df["label"]66        return X.values.tolist(), y.values.tolist()67    def get_time_per_sentence(self) -> List[int]:68        return self._df["t"].values.tolist()69@pd.api.extensions.register_dataframe_accessor("pair")70class CaPairAccessor:71    def __init__(self, df: pd.DataFrame):72        self._df = df73    def split_args_y(self) -> Tuple[List[str], List[str], List[str]]:74        args1 = self._df["arg1"].values.tolist()75        args2 = self._df["arg2"].values.tolist()76        label = self._df["label"].values.tolist()77        return args1, args2, label78    def get_time_per_sentence(self) -> List[int]:...

__init__.pyi

Source:__init__.pyi

1from pandas._libs.lib import no_default as no_default2from pandas.core.accessor import register_dataframe_accessor as register_dataframe_accessor, register_index_accessor as register_index_accessor, register_series_accessor as register_series_accessor3from pandas.core.algorithms import take as take4from pandas.core.arrays import ExtensionArray as ExtensionArray, ExtensionScalarOpsMixin as ExtensionScalarOpsMixin...

Automation Testing Tutorials

Learn to execute automation testing from scratch with LambdaTest Learning Hub. Right from setting up the prerequisites to run your first automation test, to following best practices and diving deeper into advanced test scenarios. LambdaTest Learning Hubs compile a list of step-by-step guides to help you be proficient with different test automation frameworks i.e. Selenium, Cypress, TestNG etc.