modif code

appli_test.py

@@ -2,16 +2,14 @@
 from streamlit.testing.v1 import AppTest
 from app.app import get_customers_ids, get_customer_values, get_features_selected, request_prediction, construire_jauge_score
 import matplotlib.pyplot as plt
-import pandas as pd
-#import shap
 
 @pytest.fixture
 def streamlit_client():
     script_path = "app/app.py"
 
-    # Initialize the app.
+    # Initialise l'application
     app_test = AppTest.from_file(script_path, default_timeout=3)
-    # Run the app.
+    # lance l'appli
     app_test.run()
 
     yield app_test
@@ -29,21 +27,6 @@ def test_get_features_selected(streamlit_client):
     features_selected_list = get_features_selected()
     assert isinstance(features_selected_list, list)
 
-#def test_get_customer_shap_values(streamlit_client):
-#    data_df = streamlit_client.cache(lambda: streamlit_client.head(1))()
-#    shap_values_list, _, _ = get_customer_shap_values(data_df)
-#    assert isinstance(shap_values_list, list)
-
-def test_request_prediction(streamlit_client):
-    api_url_calc = f'https://juguirlet.pythonanywhere.com/api/v1/predict'
-    #data = streamlit_client.cache(lambda: streamlit_client.head(1))()
-    mock_data = {'column_name': ['mocked_data']}
-    streamlit_client.dataframe(pd.DataFrame(mock_data))
-
-    # Now you can call your request_prediction function with the mocked data
-    response = request_prediction(api_url_calc, streamlit_client.dataframe)
-    assert "prediction" in response
-
 def test_construire_jauge_score(streamlit_client):
     score_remboursement_client = 0.7
     jauge_score = construire_jauge_score(score_remboursement_client)