stress_at_work_analysis/machine_learning/pipeline.py

import datetime
import warnings
from collections.abc import Collection
from pathlib import Path

import numpy as np
import pandas as pd
import yaml
from pyprojroot import here
from sklearn import linear_model
from sklearn.model_selection import LeaveOneGroupOut, cross_val_score

import participants.query_db
from features import communication, esm, helper, proximity
from machine_learning import QUESTIONNAIRE_IDS, QUESTIONNAIRE_IDS_RENAME

WARNING_PARTICIPANTS_LABEL = (
    "Before calculating features, please set participants label using self.set_participants_label() "
    "to be used as a filename prefix when exporting data. "
    "The filename will be of the form: %participants_label_%grouping_variable_%data_type.csv"
)


class SensorFeatures:
    def __init__(
        self,
        grouping_variable: str,
        features: dict,
        participants_usernames: Collection = None,
    ):

        self.grouping_variable_name = grouping_variable
        self.grouping_variable = [grouping_variable]

        self.data_types = features.keys()

        self.participants_label: str = ""
        if participants_usernames is None:
            participants_usernames = participants.query_db.get_usernames(
                collection_start=datetime.date.fromisoformat("2020-08-01")
            )
            self.participants_label = "all"
        self.participants_usernames = participants_usernames

        self.df_features_all = pd.DataFrame()

        self.df_proximity = pd.DataFrame()
        self.df_proximity_counts = pd.DataFrame()

        self.df_calls = pd.DataFrame()
        self.df_sms = pd.DataFrame()
        self.df_calls_sms = pd.DataFrame()

        self.folder = None
        self.filename_prefix = ""
        self.construct_export_path()
        print("SensorFeatures initialized.")

    def set_sensor_data(self):
        print("Querying database ...")
        if "proximity" in self.data_types:
            self.df_proximity = proximity.get_proximity_data(
                self.participants_usernames
            )
            print("Got proximity data from the DB.")
            self.df_proximity = helper.get_date_from_timestamp(self.df_proximity)
            self.df_proximity = proximity.recode_proximity(self.df_proximity)
        if "communication" in self.data_types:
            self.df_calls = communication.get_call_data(self.participants_usernames)
            self.df_calls = helper.get_date_from_timestamp(self.df_calls)
            print("Got calls data from the DB.")

            self.df_sms = communication.get_sms_data(self.participants_usernames)
            self.df_sms = helper.get_date_from_timestamp(self.df_sms)
            print("Got sms data from the DB.")

    def get_sensor_data(self, data_type) -> pd.DataFrame:
        if data_type == "proximity":
            return self.df_proximity
        elif data_type == "communication":
            return self.df_calls_sms
        else:
            raise KeyError("This data type has not been implemented.")

    def calculate_features(self):
        print("Calculating features ...")
        if not self.participants_label:
            raise ValueError(WARNING_PARTICIPANTS_LABEL)
        if "proximity" in self.data_types:
            self.df_proximity_counts = proximity.count_proximity(
                self.df_proximity, self.grouping_variable
            )
            self.df_features_all = safe_outer_merge_on_index(
                self.df_features_all, self.df_proximity_counts
            )
            print("Calculated proximity features.")
        to_csv_with_settings(
            self.df_proximity, self.folder, self.filename_prefix, data_type="prox"
        )

        if "communication" in self.data_types:
            self.df_calls_sms = communication.calls_sms_features(
                df_calls=self.df_calls,
                df_sms=self.df_sms,
                group_by=self.grouping_variable,
            )
            self.df_features_all = safe_outer_merge_on_index(
                self.df_features_all, self.df_calls_sms
            )
            print("Calculated communication features.")
        to_csv_with_settings(
            self.df_calls_sms, self.folder, self.filename_prefix, data_type="comm"
        )

        self.df_features_all.fillna(
            value=proximity.FILL_NA_PROXIMITY, inplace=True, downcast="infer",
        )
        self.df_features_all.fillna(
            value=communication.FILL_NA_CALLS_SMS_ALL, inplace=True, downcast="infer",
        )

    def get_features(self, data_type, feature_names) -> pd.DataFrame:
        if data_type == "proximity":
            if feature_names == "all":
                feature_names = proximity.FEATURES_PROXIMITY
            return self.df_proximity_counts[feature_names]
        elif data_type == "communication":
            if feature_names == "all":
                feature_names = communication.FEATURES_CALLS_SMS_ALL
            return self.df_calls_sms[feature_names]
        elif data_type == "all":
            return self.df_features_all
        else:
            raise KeyError("This data type has not been implemented.")

    def construct_export_path(self):
        if not self.participants_label:
            warnings.warn(WARNING_PARTICIPANTS_LABEL, UserWarning)
        self.folder = here("machine_learning/intermediate_results/features", warn=True)
        self.filename_prefix = (
            self.participants_label + "_" + self.grouping_variable_name
        )

    def set_participants_label(self, label: str):
        self.participants_label = label
        self.construct_export_path()


class Labels:
    def __init__(
        self,
        grouping_variable: list,
        labels: dict,
        participants_usernames: Collection = None,
    ):
        self.grouping_variable = grouping_variable

        self.questionnaires = labels.keys()

        if participants_usernames is None:
            participants_usernames = participants.query_db.get_usernames(
                collection_start=datetime.date.fromisoformat("2020-08-01")
            )
        self.participants_usernames = participants_usernames

        self.df_esm = pd.DataFrame()
        self.df_esm_preprocessed = pd.DataFrame()
        self.df_esm_interest = pd.DataFrame()
        self.df_esm_clean = pd.DataFrame()

        self.df_esm_means = pd.DataFrame()
        print("Labels initialized.")

    def set_labels(self):
        print("Querying database ...")
        self.df_esm = esm.get_esm_data(self.participants_usernames)
        print("Got ESM data from the DB.")
        self.df_esm_preprocessed = esm.preprocess_esm(self.df_esm)
        print("ESM data preprocessed.")
        if "PANAS" in self.questionnaires:
            self.df_esm_interest = self.df_esm_preprocessed[
                (
                    self.df_esm_preprocessed["questionnaire_id"]
                    == QUESTIONNAIRE_IDS.get("PANAS").get("PA")
                )
                | (
                    self.df_esm_preprocessed["questionnaire_id"]
                    == QUESTIONNAIRE_IDS.get("PANAS").get("NA")
                )
            ]
        self.df_esm_clean = esm.clean_up_esm(self.df_esm_interest)
        print("ESM data cleaned.")

    def get_labels(self, questionnaire):
        if questionnaire == "PANAS":
            return self.df_esm_clean
        else:
            raise KeyError("This questionnaire has not been implemented as a label.")

    def aggregate_labels(self):
        print("Aggregating labels ...")
        self.df_esm_means = (
            self.df_esm_clean.groupby(
                ["participant_id", "questionnaire_id"] + self.grouping_variable
            )
            .esm_user_answer_numeric.agg("mean")
            .reset_index()
            .rename(columns={"esm_user_answer_numeric": "esm_numeric_mean"})
        )
        self.df_esm_means = (
            self.df_esm_means.pivot(
                index=["participant_id"] + self.grouping_variable,
                columns="questionnaire_id",
                values="esm_numeric_mean",
            )
            .reset_index(col_level=1)
            .rename(columns=QUESTIONNAIRE_IDS_RENAME)
            .set_index(["participant_id"] + self.grouping_variable)
        )
        print("Labels aggregated.")

    def get_aggregated_labels(self):
        return self.df_esm_means


class ModelValidation:
    def __init__(self, X, y, group_variable=None, cv_name="loso"):
        self.model = None
        self.cv = None

        idx_common = X.index.intersection(y.index)
        self.y = y.loc[idx_common, "NA"]
        # TODO Handle the case of multiple labels.
        self.X = X.loc[idx_common]
        self.groups = self.y.index.get_level_values(group_variable)

        self.cv_name = cv_name
        print("ModelValidation initialized.")

    def set_cv_method(self):
        if self.cv_name == "loso":
            self.cv = LeaveOneGroupOut()
            self.cv.get_n_splits(X=self.X, y=self.y, groups=self.groups)
        print("Validation method set.")

    def cross_validate(self):
        print("Running cross validation ...")
        if self.model is None:
            raise TypeError(
                "Please, specify a machine learning model first, by setting the .model attribute. "
                "E.g. self.model = sklearn.linear_model.LinearRegression()"
            )
        if self.cv is None:
            raise TypeError(
                "Please, specify a cross validation method first, by using set_cv_method() first."
            )
        if self.X.isna().any().any() or self.y.isna().any().any():
            raise ValueError(
                "NaNs were found in either X or y. Please, check your data before continuing."
            )
        return cross_val_score(
            estimator=self.model,
            X=self.X,
            y=self.y,
            groups=self.groups,
            cv=self.cv,
            n_jobs=-1,
            scoring="r2",
        )


def safe_outer_merge_on_index(left, right):
    if left.empty:
        return right
    elif right.empty:
        return left
    else:
        return pd.merge(
            left,
            right,
            how="outer",
            left_index=True,
            right_index=True,
            validate="one_to_one",
        )


def to_csv_with_settings(
    df: pd.DataFrame, folder: Path, filename_prefix: str, data_type: str
) -> None:
    export_filename = filename_prefix + "_" + data_type + ".csv"
    full_path = folder / export_filename
    df.to_csv(
        path_or_buf=full_path,
        sep=",",
        na_rep="NA",
        header=True,
        index=False,
        encoding="utf-8",
    )
    print("Exported the dataframe to " + str(full_path))


if __name__ == "__main__":
    with open("./config/prox_comm_PANAS_features.yaml", "r") as file:
        sensor_features_params = yaml.safe_load(file)
    sensor_features = SensorFeatures(**sensor_features_params)
    sensor_features.set_sensor_data()
    sensor_features.calculate_features()

    with open("./config/prox_comm_PANAS_labels.yaml", "r") as file:
        labels_params = yaml.safe_load(file)
    labels = Labels(**labels_params)
    labels.set_labels()
    labels.aggregate_labels()

    model_validation = ModelValidation(
        sensor_features.get_features("all", "all"),
        labels.get_aggregated_labels(),
        group_variable="participant_id",
        cv_name="loso",
    )
    model_validation.model = linear_model.LinearRegression()
    model_validation.set_cv_method()
    model_loso_r2 = model_validation.cross_validate()
    print(model_loso_r2)
    print(np.mean(model_loso_r2))