Menggunakan variabel evolusi

Predictive Analytics Tingkat Menengah dengan Python

Nele Verbiest

Senior Data Scientist @PythonPredictions

Membangun model prediktif

# Import modul linear_model
from sklearn import linear_model

# Variabel prediktor
variables = ["gender","age", "donations_last_year", "ratio_month_year"]

# Pilih prediktor dan target
X = basetable[variables]
y = basetable[["target"]]

# Bangun model regresi logistik
logreg = linear_model.LogisticRegression()
logreg.fit(X, y)

Membuat prediksi

# Import modul linear_model
from sklearn import linear_model

# Variabel prediktor
variables = ["gender","age", "donations_last_year", "ratio_month_year"]

# Pilih prediktor dan target
X = basetable[variables]
y = basetable[["target"]]

# Bangun model regresi logistik
logreg = linear_model.LogisticRegression()
logreg.fit(X, y)

# Buat prediksi
predictions = logreg.predict_proba(X)[:,1]

Evaluasi model prediktif dengan AUC

# Impor roc_auc_score dari sklearn.metrics
from sklearn.metrics import roc_auc_score

# Hitung AUC
auc= roc_auc_score(y, predictions)
print(round(auc,2))

0.56

Predictor insight graph

# Diskretkan variabel menjadi 5 bin dan tambahkan ke basetable
basetable["ratio_month_year_disc"] = pd.qcut(basetable["ratio_month_year"], 5)

# Bangun tabel predictor insight graph
pig_table = create_pig_table(basetable, "target","ratio_month_year_disc")

```{python}
# Plot predictor insight graph
plot_pig(pig_table, "ratio_month_year_disc")

Interpretasi predictor insight graph

Ayo berlatih!

Predictive Analytics Tingkat Menengah dengan Python