01Evaluation.py

import pandas as pd 
from typing import List, Optional
import math

def linearRegressionScore(X, y):
    # using linear regression score function to compute the r2
    from sklearn.linear_model import LinearRegression
    model = LinearRegression()
    model.fit(X, y)
    r_squared = model.score(X, y)
    return r_squared

def evaluation(predictionAge):
    from sklearn.metrics import mean_absolute_error, mean_squared_error
    mean_squared_error = mean_squared_error(predictionAge["Age"].values, predictionAge["prediction"].values)
    mean_absolute_error = mean_absolute_error(predictionAge["Age"].values, predictionAge["prediction"].values)
    r2_score = linearRegressionScore(predictionAge[["Age"]], predictionAge[["prediction"]])
    dataCorr = predictionAge["Age"].corr(predictionAge['prediction'], method='pearson')
    return mean_squared_error, mean_absolute_error, r2_score, dataCorr


SkinPathwayAgePrecdiction = pd.read_pickle("./Precdiction.pickle")

MSEList = []
MAEList = []
R2List = []
RhoList = []
CohortList = []

for cohort in SkinPathwayAgePrecdiction.Cohort.unique():
    # test the control group only
    predictionPerCohort = SkinPathwayAgePrecdiction[SkinPathwayAgePrecdiction.Label.eq(0)  & SkinPathwayAgePrecdiction.Cohort.eq(cohort)].dropna()
    if not predictionPerCohort.empty:
        mean_squared_error, mean_absolute_error, r2_score, dataCorr = evaluation(predictionPerCohort)
        MSEList.append(mean_squared_error)
        MAEList.append(mean_absolute_error)
        R2List.append(r2_score)
        RhoList.append(dataCorr)
        CohortList.append(cohort)

Evaluation = pd.DataFrame(data = [MSEList, MAEList, R2List, RhoList, CohortList], index = ["MSE", "MAE", "r2", "Corr", "Cohort"]).T
Evaluation["Tag"]= "Control"
Evaluation