Mashq 3: yashirin leakage

import pandas as pd

td = pd.read_csv(DATA / "talaba_bitirdi.csv", keep_default_na=False, na_values=[""])
print("shakli:", td.shape)
print()
print(td.dtypes.to_string())
print()
show(td.head(8))

print("Target:")
print(td.bitirdi.value_counts(normalize=True).sort_index().round(4).to_string())
print()
print(f"BAZA (hammaga '1' desa): {td.bitirdi.mean():.4f}")
print()
yoq = td.isna().sum()
print("Bo'sh qiymatlar:")
print(yoq[yoq > 0].to_string() if (yoq > 0).any() else "  yo'q")

import pandas as pd
show(pd.DataFrame([
    {"#": 1, "Yozib qo'yiladigan javob": "Nechta qator, nechta ustun"},
    {"#": 2, "Yozib qo'yiladigan javob": "\"Men ___ ni ___ dan bashorat qilaman\" — bitta jumla"},
    {"#": 3, "Yozib qo'yiladigan javob": "Vazifa turi: regression / binary / multiclass"},
    {"#": 4, "Yozib qo'yiladigan javob": "Target nisbati va BAZA raqami"},
    {"#": 5, "Yozib qo'yiladigan javob": "Split qildingizmi? stratify kerakmi?"},
    {"#": 6, "Yozib qo'yiladigan javob": "Dummy natijasi"},
    {"#": 7, "Yozib qo'yiladigan javob": "Qaysi ustunlarni tashladingiz va NEGA"},
    {"#": 8, "Yozib qo'yiladigan javob": "Missing va turlar bilan nima qildingiz"},
    {"#": 9, "Yozib qo'yiladigan javob": "Birinchi halol natijangiz"},
    {"#": 10, "Yozib qo'yiladigan javob": "Qaysi metrikani tanladingiz va NEGA"},
    {"#": 11, "Yozib qo'yiladigan javob": "Xatolar qaysi guruhda to'plangan"},
    {"#": 12, "Yozib qo'yiladigan javob": "Qaysi BITTA o'zgarishni qildingiz va natija"},
]))

from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression
from sklearn.dummy import DummyClassifier
from sklearn.metrics import accuracy_score, f1_score, recall_score, precision_score

halol = ["kirish_bali", "davomat_foiz", "uy_vazifa_bali", "ishlaydi", "masofa_km"]
y = td["bitirdi"]
tr, te, ytr, yte = train_test_split(td, y, test_size=0.3, random_state=42, stratify=y)

baza = accuracy_score(yte, DummyClassifier(strategy="most_frequent").fit(tr[halol], ytr).predict(te[halol]))
m1 = LogisticRegression(max_iter=2000).fit(tr[halol], ytr)
b1 = m1.predict(te[halol])

print(f"BAZA                    : {baza:.4f}")
print(f"Halol 5 ustun — accuracy: {accuracy_score(yte, b1):.4f}   (farq {accuracy_score(yte, b1)-baza:+.4f})")
print(f"Halol 5 ustun — F1      : {f1_score(yte, b1):.4f}")
print(f"Halol 5 ustun — recall  : {recall_score(yte, b1):.4f}")

td2 = td.copy()
td2["diplom_bor"] = td2.diplom_seriya.notna().astype(int)
tr2, te2 = td2.loc[tr.index], td2.loc[te.index]

m2 = LogisticRegression(max_iter=2000).fit(tr2[halol + ["diplom_bor"]], ytr)
b2 = m2.predict(te2[halol + ["diplom_bor"]])
print(f"+ diplom_seriya (bor/yo'q): accuracy {accuracy_score(yte, b2):.4f}   <-- MUKAMMAL")
print()
print("Nega? Tekshiramiz:")
print(pd.crosstab(td2.diplom_bor, td2.bitirdi).to_string())

m3 = LogisticRegression(max_iter=2000).fit(tr[halol + ["oxirgi_tolov_oyi"]], ytr)
b3 = m3.predict(te[halol + ["oxirgi_tolov_oyi"]])
print(f"+ oxirgi_tolov_oyi: accuracy {accuracy_score(yte, b3):.4f}  F1 {f1_score(yte, b3):.4f}")
print()
print("Guruhlar bo'yicha o'rtacha:")
print(td.groupby("bitirdi").oxirgi_tolov_oyi.agg(["mean", "min", "max"]).round(2).to_string())

show(pd.DataFrame([
    {"Ustun": "diplom_seriya",
     "Target bilan bog'liq": "To'liq (100%)",
     "Bashorat paytida bormi": "YO'Q — bitirgandan keyin beriladi",
     "Hukmi": "LEAKAGE — tashlanadi"},
    {"Ustun": "oxirgi_tolov_oyi",
     "Target bilan bog'liq": "Qisman",
     "Bashorat paytida bormi": "HA — har oy ma'lum bo'ladi",
     "Hukmi": "Haqiqiy signal — ishlatiladi (lekin ehtiyot bilan)"},
]))

show(pd.DataFrame([
    {"Mashq": "42. penguins", "Nimani sinadi": "Naqsh buzilishi — binary dan multiclass ga",
     "Asosiy dars": "average='macro' va 3x3 matritsa"},
    {"Mashq": "43. iflos savdo", "Nimani sinadi": "Tozalash ishning 80 foizi",
     "Asosiy dars": "Kirill/lotin, matn son, dublikat; va 'model qurib bo'lmaydi' halol xulosasi"},
    {"Mashq": "44. yashirin leakage", "Nimani sinadi": "Tuzoqni o'zingiz topish",
     "Asosiy dars": "Leakage ni VAQT savoli ajratadi, bog'liqlik kuchi emas"},
]))