Forex Forecasting with Random Forests – 1/N Preproccesing

# Import libraries

import numpy as np
import pandas as pd
import datetime as dt

from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import accuracy_score
import optuna

from sklearn.metrics import confusion_matrix
# from sklearn.metrics import plot_confusion_matrix
from sklearn.metrics import accuracy_score, confusion_matrix, ConfusionMatrixDisplay

import seaborn as sns
import matplotlib.pyplot as plt
import matplotlib.dates as mdates
import matplotlib as mpl
import matplotlib.ticker as ticker

# How to install talib https://tkstock.site/2018/08/22/post-593/
import talib as ta

# Evaluation function

def win_loss(Y_test, pred):
    # Data framing of Series type data
    test_y2 = pd.DataFrame(Y_test)
    # Store forecast results
    test_y2['pred'] = pred

    # Define each cell of the confusion matrix
    m1 = len(test_y2[(test_y2['Maguro']==1) & (test_y2['pred']==1)])
    m2 = len(test_y2[(test_y2['Maguro']==1) & (test_y2['pred']==0)])
    m3 = len(test_y2[(test_y2['Maguro']==1) & (test_y2['pred']==-1)])
    m4 = len(test_y2[(test_y2['Maguro']==0) & (test_y2['pred']==1)])
    m5 = len(test_y2[(test_y2['Maguro']==0) & (test_y2['pred']==0)])
    m6 = len(test_y2[(test_y2['Maguro']==0) & (test_y2['pred']==-1)])
    m7 = len(test_y2[(test_y2['Maguro']==-1) & (test_y2['pred']==1)])
    m8 = len(test_y2[(test_y2['Maguro']==-1) & (test_y2['pred']==0)])
    m9 = len(test_y2[(test_y2['Maguro']==-1) & (test_y2['pred']==-1)])

    #　Calculation of total number of transactions, percentage of large wins, percentage of large losses, and
    try:
        mall=m1+m2+m3+m4+m5+m6+m7+m8+m9
        ma=m1+m3+m4+m6+m7+m9
        mb=(m1+m9)/ma
        mc=(m3+m7)/ma
        win_loss_ratio = mb/mc
    except ZeroDivisionError:
        win_loss_ratio = 0
        
    # Output of various model evaluation indices
    return win_loss_ratio