From 53e90284644d1e361f87d450b3d1767abd9e5616 Mon Sep 17 00:00:00 2001
From: slowy07 <slowy.arfy@gmail.com>
Date: Sat, 15 Apr 2023 13:47:48 +0700
Subject: [PATCH 1/2] feat: menambahkan fungsi skoring

---
 .../artificial_intelligence/fungsi_skoring.py | 97 +++++++++++++++++++
 1 file changed, 97 insertions(+)
 create mode 100644 implementation/artificial_intelligence/fungsi_skoring.py

diff --git a/implementation/artificial_intelligence/fungsi_skoring.py b/implementation/artificial_intelligence/fungsi_skoring.py
new file mode 100644
index 00000000..796bab39
--- /dev/null
+++ b/implementation/artificial_intelligence/fungsi_skoring.py
@@ -0,0 +1,97 @@
+# implementasi dari fungsi skoring
+# untuk mengkalkulasikan perbedaan antara
+# value yang terprediksi dengan value yang aktual
+# terdapat juga mode dari:
+# - MAE (mean absolute error)
+#   menghitung perbedaan mutlak rata-rata antara
+#   dua vektor numerik
+# - RMSE (root mean squared error)
+#   untuk mengukur seberapa jauh nilai observasi
+#   dalam analisis regresi
+# - MSE (mean squared error)
+#   fungsi untuk mengukur square error
+# - RMSLE (root mean square logarithmic error)
+#   menambahkan satu ke aktual sebelum mengambil
+#   nilai logaritma natural untuk menghindari
+#   catching log natural nilai nol
+
+import numpy as np
+
+
+def fungsi_mae(prediksi: list, nilai_aktual: list) -> float:
+    """
+    >>> nilai_aktual = [1,2,3]
+    >>> prediksi = [1, 4, 3]
+    >>> np.around(fungsi_mae(prediksi, nilai_aktual), decimals=2)
+    0.67
+    """
+    prediksi = np.array(prediksi)
+    nilai_aktual = np.array(nilai_aktual)
+    perbedaan = abs(prediksi - nilai_aktual)
+    hasil = perbedaan.mean()
+    return hasil
+
+
+def fungsi_mse(prediksi: list, nilai_aktual: list) -> float:
+    """
+    >>> nilai_aktual = [1, 1, 1]
+    >>> prediksi = [1, 1, 1]
+    >>> fungsi_mse(prediksi, nilai_aktual)
+    0.0
+    """
+    prediksi = np.array(prediksi)
+    nilai_aktual = np.array(nilai_aktual)
+    perbedaan = prediksi - nilai_aktual
+    root_diff = np.square(perbedaan)
+    hasil = root_diff.mean()
+    return hasil
+
+
+def fungsi_rmse(prediksi: list, nilai_aktual: list) -> float:
+    """
+    >>> nilai_aktual = [1, 2, 3]
+    >>> prediksi = [1, 4, 3]
+    >>> np.around(fungsi_rmse(prediksi, nilai_aktual), decimals=2)
+    1.15
+    """
+    prediksi = np.array(prediksi)
+    nilai_aktual = np.array(nilai_aktual)
+    perbedaan = prediksi - nilai_aktual
+    root_diff = np.square(perbedaan)
+    mean_square_diff = root_diff.mean()
+    hasil = np.sqrt(mean_square_diff)
+    return hasil
+
+
+def fungsi_rmsle(prediksi: list, nilai_aktual: list) -> float:
+    """
+    >>> nilai_aktual = [10, 10, 30]
+    >>> prediksi = [10, 2, 30]
+    >>> np.around(fungsi_rmsle(prediksi, nilai_aktual), decimals = 2)
+    0.75
+    """
+    prediksi = np.array(prediksi)
+    nilai_aktual = np.array(nilai_aktual)
+    prediksi_log = np.log(prediksi + 1)
+    nilai_log_aktual = np.log(nilai_aktual + 1)
+    perbedaan = prediksi_log - nilai_log_aktual
+    square_diff = np.square(perbedaan)
+    mean_square_diff = square_diff.mean()
+    hasil = np.sqrt(mean_square_diff)
+    return hasil
+
+
+def fungsi_mbd(prediksi: list, nilai_aktual: list) -> float:
+    prediksi = np.array(prediksi)
+    nilai_aktual = np.array(nilai_aktual)
+    perbedaan = prediksi - nilai_aktual
+    numerator = np.sum(perbedaan) / len(prediksi)
+    denumerator = np.sum(nilai_aktual) / len(prediksi)
+    hasil = float(numerator) / denumerator * 100
+    return hasil
+
+
+if __name__ == "__main__":
+    import doctest
+
+    doctest.testmod(verbose=True)

From f5edb2e4ad98f7261cd860154ea496e02f9b1f4f Mon Sep 17 00:00:00 2001
From: github-actions <bellshdae07@gmail.com>
Date: Sat, 15 Apr 2023 06:48:18 +0000
Subject: [PATCH 2/2] update DIRECTORY.md

---
 DIRECTORY.md | 1 +
 1 file changed, 1 insertion(+)

diff --git a/DIRECTORY.md b/DIRECTORY.md
index f66d6624..40de3f7d 100644
--- a/DIRECTORY.md
+++ b/DIRECTORY.md
@@ -148,6 +148,7 @@
 ## Implementation
   * Artificial Intelligence
     * [Decision Tree](https://github.com/bellshade/Python/blob/main/implementation/artificial_intelligence/decision_tree.py)
+    * [Fungsi Skoring](https://github.com/bellshade/Python/blob/main/implementation/artificial_intelligence/fungsi_skoring.py)
     * [Gradient Descent](https://github.com/bellshade/Python/blob/main/implementation/artificial_intelligence/gradient_descent.py)
     * [K Means Clutser](https://github.com/bellshade/Python/blob/main/implementation/artificial_intelligence/k_means_clutser.py)
     * [K Nearest Neighbour](https://github.com/bellshade/Python/blob/main/implementation/artificial_intelligence/k_nearest_neighbour.py)