app

Dockerfile

@@ -0,0 +1,16 @@
+FROM python:3.12.4-slim
+
+WORKDIR /app
+
+COPY app/ /app/
+COPY pyproject.toml /app
+
+RUN pip install --no-cache-dir poetry
+
+RUN poetry install --no-root
+
+EXPOSE 7860
+
+ENV GRADIO_SERVER_NAME="0.0.0.0"
+
+CMD ["poetry", "run", "python", "gradio_app.py"]

app/.gradio/flagged/dataset1.csv

@@ -0,0 +1,3 @@
+TotalHouseArea,OverallQual,NeighborhoodPrice,GrLivArea,GarageEfficency,BuildingAge,KitchenQual,BsmtQual,GarageAge,TotalFullBath,RemodAge,GarageFinish,ExterScore,LotFrontage,TotRmsAbvGrd,BathPerBedroom,SalePrice,Predicted Value,timestamp
+,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,,2024-12-17 15:05:29.544565
+,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,,2024-12-17 15:05:35.128059

app/gradio_app.py

@@ -0,0 +1,183 @@
+import gradio as gr
+import pandas as pd
+import joblib
+from pathlib import Path
+
+
+def denormalize(data):
+    scaler = joblib.load("scaler.joblib")
+    data_denorm = scaler.inverse_transform(data)
+    denormalized_df = pd.DataFrame(data_denorm, columns=scaler.feature_names_in_)
+    return denormalized_df
+
+
+def prepare_for_predict(*inputs):
+    scaler = joblib.load("scaler.joblib")
+
+    scaler_columns = scaler.feature_names_in_
+
+    predict_data = pd.DataFrame([inputs], columns=scaler_columns)
+
+    kitchen_qual_mapping = {
+        "Doskonała": 0,
+        "Dobra": 2,
+        "Ponadprzeciętna": 3,
+        "Przeciętna": 1,
+    }
+
+    bsmt_qual_mapping = {
+        "Doskonała": 0,
+        "Dobra": 2,
+        "Ponadprzeciętna": 3,
+        "Przeciętna": 1,
+    }
+
+    garage_finish_mapping = {
+        "Prawie wykończony": 1,
+        "Niewykończony": 2,
+        "Wykończony": 0,
+    }
+
+    exter_score_mapping = {
+        "Doskonała": 10,
+        "Bardzo dobra": 9,
+        "Dobra": 8,
+        "Ponadprzeciętna": 7,
+        "Przeciętna": 6,
+        "Słaba": 5,
+        "Bardzo słaba": 4,
+    }
+
+    predict_data["KitchenQual"] = kitchen_qual_mapping.get(
+        predict_data["KitchenQual"][0], None
+    )
+    predict_data["BsmtQual"] = bsmt_qual_mapping.get(predict_data["BsmtQual"][0], None)
+    predict_data["GarageFinish"] = garage_finish_mapping.get(
+        predict_data["GarageFinish"][0], None
+    )
+    predict_data["ExterScore"] = exter_score_mapping.get(
+        predict_data["ExterScore"][0], None
+    )
+
+    predict_data = scaler.transform(predict_data)
+
+    return predict_data
+
+
+def predict(*inputs):
+    model = joblib.load("model1.joblib")
+
+    predict_data = prepare_for_predict(*inputs)
+
+    prediction = model.predict(predict_data)
+
+    return f"{prediction[0]:.2f} USD."
+
+
+def app():
+    data = pd.read_csv(Path("prepared_train.csv"))
+    data = data.drop(["SalePrice"], axis=1)
+
+    columns = data.columns.tolist()
+
+    inputs = []
+
+    original_values = denormalize(data)
+    mean_values = original_values.mean()
+
+    feature_translation = {
+        "TotalHouseArea": "Całkowita powierzchnia domu w stopach kwadratowych",
+        "OverallQual": "Ogólna jakość budynku w skali od 0 do 10",
+        "NeighborhoodPrice": "Średnia cena w sąsiedztwie",
+        "GrLivArea": "Powierzchnia mieszkalna nad poziomem gruntu w stopach kwadratowych",
+        "GarageEfficency": "Efektywność garażu",
+        "BuildingAge": "Wiek budynku",
+        "KitchenQual": "Jakość kuchni",
+        "BsmtQual": "Jakość piwnicy",
+        "GarageAge": "Wiek garażu",
+        "TotalFullBath": "Liczba pełnych łazienek",
+        "RemodAge": "Wiek ostatniego remontu",
+        "GarageFinish": "Wykończenie garażu",
+        "ExterScore": "Ocena stanu zewnętrznego",
+        "LotFrontage": "Licznik metrów bieżących ulicy połączonej z nieruchomością",
+        "TotRmsAbvGrd": "Ilość pokoi ponad poziomem gruntu",
+        "BathPerBedroom": "Stosunek łązienek do sypialni",
+    }
+
+    for col in columns:
+        if col == "KitchenQual":
+            inputs.append(
+                gr.Dropdown(
+                    label=f"{col} ({feature_translation.get(col, 'Brak tłumaczenia')})",
+                    choices=["Doskonała", "Dobra", "Ponadprzeciętna", "Przeciętna"],
+                    value="Dobra",
+                )
+            )
+        elif col == "BsmtQual":
+            inputs.append(
+                gr.Dropdown(
+                    label=f"{col} ({feature_translation.get(col, 'Brak tłumaczenia')})",
+                    choices=["Doskonała", "Dobra", "Ponadprzeciętna", "Przeciętna"],
+                    value="Dobra",
+                )
+            )
+        elif col == "OverallQual":
+            inputs.append(
+                gr.Dropdown(
+                    label=f"{col} ({feature_translation.get(col, 'Brak tłumaczenia')})",
+                    choices=[10, 9, 8, 7, 6, 5, 4, 3, 2, 1],
+                    value=5,
+                )
+            )
+        elif col == "GarageFinish":
+            inputs.append(
+                gr.Dropdown(
+                    label=f"{col} ({feature_translation.get(col, 'Brak tłumaczenia')})",
+                    choices=["Prawie wykończony", "Niewykończony", "Wykończony"],
+                    value="Wykończony",
+                )
+            )
+        elif col == "ExterScore":
+            inputs.append(
+                gr.Dropdown(
+                    label=f"{col} ({feature_translation.get(col, 'Brak tłumaczenia')})",
+                    choices=[
+                        "Doskonała",
+                        "Bardzo dobra",
+                        "Dobra",
+                        "Ponadprzeciętna",
+                        "Przeciętna",
+                        "Słaba",
+                        "Bardzo słaba",
+                    ],
+                    value="Dobra",
+                )
+            )
+        else:
+            mean_value = round(mean_values.get(col, 0), 0)
+            inputs.append(
+                gr.Number(
+                    label=f"{col} ({feature_translation.get(col, 'Brak tłumaczenia')})",
+                    value=mean_value,
+                )
+            )
+
+    with gr.Blocks() as iface:
+        gr.Markdown("# Prognozowanie cen nieruchomości")
+        with gr.Row():
+            for i in range(0, len(inputs), 16):
+                for j in range(4):
+                    with gr.Column():
+                        for input_field in inputs[i + j * 4 : i + (j + 1) * 4]:
+                            input_field.render()
+
+        predict_button = gr.Button("Prognozuj")
+        output = gr.Textbox(label="Przewidziana wartość")
+
+        predict_button.click(fn=predict, inputs=inputs, outputs=output)
+
+    iface.launch()
+
+
+if __name__ == "__main__":
+    app()

app/model1.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7fc752efab0870031db12763adef25236e9a7464d28c7dff36d6725facffff1d
+size 122765461

app/scaler.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:addd619c43d56bcff67128c5cec4038ea42b1d8aaf642e0739f64c9978df4bad
+size 1895

pyproject.toml

@@ -0,0 +1,29 @@
+[tool.poetry]
+name = "praca-dyplomowa"
+version = "0.2.0"
+description = ""
+authors = ["Perunio <divodar@protonmail.com>"]
+readme = "README.md"
+
+[tool.poetry.dependencies]
+python = "^3.12"
+scikit-learn = "^1.5.2"
+scipy = "^1.14.1"
+matplotlib = "^3.9.2"
+seaborn = "^0.13.2"
+ruff = "^0.7.4"
+pytest = "^8.3.3"
+optuna = {version = "^4.1.0", extras = ["dashboard"]}
+optuna-dashboard = "^0.17.0"
+aiohttp = "^3.11.10"
+jinja2 = "^3.1.4"
+
+pandas = "^2.2.3"
+xgboost = "^2.1.3"
+joblib = "^1.4.2"
+gradio = "^5.9.1"
+
+
+[build-system]
+requires = ["poetry-core"]
+build-backend = "poetry.core.masonry.api"