Add Car Purchase Simulator files

.gradio/flagged/dataset1.csv

@@ -0,0 +1,6 @@
+Car Type,Interest Cut (%),Simulation Results,timestamp
+1,0.01,"{""headers"": [""Car Type"", ""Current Interest Rate (%)"", ""Current Purchase Probability"", ""Current Expected Purchase Amount"", ""New Interest Rate (%)"", ""New Purchase Probability"", ""New Expected Purchase Amount""], ""data"": [[1, 3.9, 0.3, 46.49, 3.89, 0.3002, 46.53]], ""metadata"": null}",2025-02-23 01:10:08.407855
+2,0.5,"{""headers"": [""Car Type"", ""Current Interest Rate (%)"", ""Current Purchase Probability"", ""Current Expected Purchase Amount"", ""New Interest Rate (%)"", ""New Purchase Probability"", ""New Expected Purchase Amount""], ""data"": [[2, 3.9, 0.3281, 47.23, 3.4, 0.3469, 49.94]], ""metadata"": null}",2025-02-23 01:11:23.857094
+2,0.5,"{""headers"": [""Car Type"", ""Current Interest Rate (%)"", ""Current Purchase Probability"", ""Current Expected Purchase Amount"", ""New Interest Rate (%)"", ""New Purchase Probability"", ""New Expected Purchase Amount""], ""data"": [[2, 3.9, 0.3281, 47.23, 3.4, 0.3469, 49.94]], ""metadata"": null}",2025-02-23 01:11:27.432303
+2,0.5,"{""headers"": [""Car Type"", ""Current Interest Rate (%)"", ""Current Purchase Probability"", ""Current Expected Purchase Amount"", ""New Interest Rate (%)"", ""New Purchase Probability"", ""New Expected Purchase Amount""], ""data"": [[2, 3.9, 0.3281, 47.23, 3.4, 0.3469, 49.94]], ""metadata"": null}",2025-02-23 01:11:29.984112
+2,0.5,"{""headers"": [""Car Type"", ""Current Interest Rate (%)"", ""Current Purchase Probability"", ""Current Expected Purchase Amount"", ""New Interest Rate (%)"", ""New Purchase Probability"", ""New Expected Purchase Amount""], ""data"": [[2, 3.9, 0.3281, 47.23, 3.4, 0.3469, 49.94]], ""metadata"": null}",2025-02-23 01:11:41.982872

20250221_Car_LogitData.csv

@@ -0,0 +1,11 @@
+x1,x2,x3,x4,x5,x6,x7,x8,x9,x10,x11
+154.9660956,3.9,1.9,0.364,0.45,0.404567463,0.606851194,1.011418657,1.41598612,-0.397073136,-0.115442237
+143.9424424,3.9,1.9,0.519,0.589,0.453831298,0.680746948,1.134578246,1.588409545,-0.060856621,-0.168159245
+173.1029233,3.9,1.9,0.473,0.554,0.391369889,0.587054834,0.978424723,1.369794612,-0.120264206,-0.211742269
+174.6740746,3.9,2.5,0.484,0.625,0.326957931,0.490436897,0.817394828,1.144352759,0.801816352,-0.406712703
+216.1393818,3.9,1.9,0.445,0.55,0.327608294,0.491412441,0.819020736,1.14662903,0.220874641,-0.263217945
+246.7452458,3.9,2.9,0.482,0.592,0.286523272,0.429784908,0.716308179,1.002831451,1.45423119,-0.564521768
+296.1389772,3.9,2.5,0.479,0.516,0.222169897,0.333254846,0.555424743,0.777594641,0.112258804,-0.209057934
+337.2407599,4.3,1.9,0.524,0.741,0.194013122,0.291019684,0.485032806,0.679045929,1.039755788,-0.334110966
+463.8277575,4.3,1.9,0.411,0.651,0.158982591,0.238473886,0.397456476,0.556439067,0.605771068,-0.296186149
+363.3534513,3.9,0.9,0.444,0.771,0.179777895,0.269666843,0.449444739,0.629222634,1.136518144,-0.469911503

Car_Purchase_Results.csv

@@ -0,0 +1,5 @@
+Car_Type,Condition,Interest_Rate,Purchase_Probability,Expected_Purchase_Amount
+8,Current,-7,0.9670268462555028,1934055.6265646983
+8,After Cut (-2%),0,0.738802882465617,1477607.2425369988
+1,Current,3.9,0.3000000000183128,600000.0000366255
+1,After Cut (0.5%),3.4,0.31225955838462643,624519.1167692529

app.py

@@ -0,0 +1,81 @@
+import gradio as gr
+import pandas as pd
+import numpy as np
+
+# データの読み込み
+data = pd.read_csv("20250221_Car_LogitData.csv")
+
+# 結果を格納するデータフレーム
+results_df = pd.DataFrame(columns=["Car_Type", "Condition", "Interest_Rate", "Purchase_Probability", "Expected_Purchase_Amount"])
+
+# ロジット関数の定義
+def logistic(x, beta_0, beta_1):
+    return 1 / (1 + np.exp(-(beta_0 + beta_1 * x)))
+
+# 分析関数
+def perform_analysis(car_type, avg_price, interest_rate, cut_rate):
+    global results_df  # 既存の結果に追加するためにグローバル変数を利用
+
+    # 入力値の制限チェック
+    if not (1 <= car_type <= 10):
+        return "Error: Car_Type must be between 1 and 10."
+    if interest_rate < 0 or cut_rate < 0:
+        return "Error: Interest rates must be non-negative."
+
+    # CSVから切片と係数を取得
+    beta_0 = data.iloc[int(car_type) - 1, 9]  # 10列目（0-indexなので9）
+    beta_1 = data.iloc[int(car_type) - 1, 10] # 11列目（0-indexなので10）
+
+    # 現在の購入確率と期待購入額
+    purchase_prob_current = logistic(interest_rate, beta_0, beta_1)
+    expected_purchase_current = purchase_prob_current * avg_price
+
+    # 金利を引き下げた場合の計算（0%未満にならないように制限）
+    new_interest_rate = max(0, interest_rate - cut_rate)
+    purchase_prob_new = logistic(new_interest_rate, beta_0, beta_1)
+    expected_purchase_new = purchase_prob_new * avg_price
+
+    # 結果を保存
+    new_result = pd.DataFrame({
+        "Car_Type": [car_type, car_type],
+        "Condition": ["Current", f"After Cut ({cut_rate}%)"],
+        "Interest_Rate": [interest_rate, new_interest_rate],
+        "Purchase_Probability": [purchase_prob_current, purchase_prob_new],
+        "Expected_Purchase_Amount": [expected_purchase_current, expected_purchase_new]
+    })
+
+    results_df = pd.concat([results_df, new_result], ignore_index=True)
+
+    return new_result
+
+# CSVダウンロード関数
+def download_results():
+    csv_path = "Car_Purchase_Results.csv"
+    results_df.to_csv(csv_path, index=False)
+    return csv_path
+
+# Gradio インターフェースの作成
+with gr.Blocks() as demo:
+    gr.Markdown("### 🚗 車種ごとの金利と購入確率の分析ツール（CSVデータからβ0とβ1を自動取得）")
+
+    with gr.Row():
+        car_type = gr.Slider(label="Car Type (1-10)", minimum=1, maximum=10, step=1, value=1)
+        avg_price = gr.Number(label="Average Purchase Price", value=2000000)
+        interest_rate = gr.Number(label="Current Interest Rate (%)", value=3.9, minimum=0)
+        cut_rate = gr.Number(label="Interest Rate Cut (%)", value=0.5, minimum=0)
+
+    output = gr.Dataframe(label="Analysis Result")
+
+    analyze_button = gr.Button("Run Analysis")
+    analyze_button.click(
+        perform_analysis,
+        inputs=[car_type, avg_price, interest_rate, cut_rate],
+        outputs=output
+    )
+
+    download_button = gr.Button("Download Results as CSV")
+    download_file = gr.File()
+    download_button.click(download_results, outputs=download_file)
+
+# 実行
+demo.launch()

requirements.txt

@@ -0,0 +1,3 @@
+gradio
+pandas
+numpy