Develop

.github/workflows/test.yml

@@ -21,6 +21,7 @@ jobs:
       run: |
         python -m pip install --upgrade pip
         pip install pytest great_expectations pandas scikit-learn flake8 black mypy pytest-cov
+        pip install mlflow numpy tensorflow
         if [ -f requirements.txt ]; then pip install -r requirements.txt; fi
 
     - name: Lint with flake8
@@ -39,3 +40,13 @@ jobs:
     - name: Run model tests
       run: |
         pytest day5/演習3/tests/test_model.py -v
+
+    # 追加: モデルの推論精度・時間の検証
+    - name: Run model inference tests
+      run: |
+        pytest day5/演習3/tests/test_model_inference.py -v
+
+    # 追加: モデルのパフォーマンス比較テスト
+    - name: Run model performance comparison
+      run: |
+        pytest day5/演習3/tests/test_model_comparison.py -v

day5/演習2/black_check.py

@@ -1,7 +1,18 @@
+def say_hello(name):
+    print("Hello," + name + "!")  # greet
+
+
+def say_hello(name):
+    print("Hello," + name + "!")  # greet
+
+
+def add(a, b):
+    return a + b
+
+
+def add(a, b):
+    return a + b
+
 
-def say_hello(name):print("Hello,"+name+"!")   # greet
-def say_hello(name):print("Hello,"     +    name   +"!")   # greet
-def add( a,b):return a+b
-def add( a ,     b    ):return a+b
 def add(a, b):
-    return a+b
+    return a + b

day5/演習2/main.py

@@ -11,6 +11,7 @@
 import time
 import great_expectations as gx
 
+
 class DataLoader:
     """データロードを行うクラス"""
 

day5/演習3/tests/test_model_inference.py

@@ -0,0 +1,43 @@
+import pytest
+import time
+import json
+import os
+import pickle
+import numpy as np
+import pandas as pd
+from pathlib import Path
+
+def load_model():
+    """モデルをロードする関数"""
+    # day5/演習1 のモデルを使用
+    model_path = os.path.join("day5", "演習1", "models", "titanic_model.pkl")
+    with open(model_path, "rb") as f:
+        model = pickle.load(f)
+    return model
+
+def load_test_data():
+    """テストデータをロードする関数"""
+    # day5/演習1 のデータを使用
+    data_path = os.path.join("day5", "演習1", "data", "test.csv")
+
+    # データを読み込む
+    df = pd.read_csv(data_path)
+
+    # 前処理
+    if 'Survived' in df.columns:
+        X = df.drop(['Survived'], axis=1)
+        y = df['Survived']
+    else:
+        X = df
+        # ダミーのラベル
+        y = pd.Series([0, 1] * (len(df) // 2) + [0] * (len(df) % 2))
+
+    # カテゴリカル変数の処理
+    X = pd.get_dummies(X)
+
+    return X, y
+
+def test_model_inference_accuracy():
+    """モデルの推論精度をテスト"""
+    model = load_model()
+    X, y = load_test_data()