Update app.py

app.py

@@ -1,89 +1,67 @@
-import streamlit as st
 import pandas as pd
-import matplotlib.pyplot as plt
-from datasets import load_dataset
+import numpy as np
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import StandardScaler, OneHotEncoder
+from sklearn.compose import ColumnTransformer
+from sklearn.pipeline import Pipeline
+from sklearn.ensemble import RandomForestRegressor
+from sklearn.metrics import mean_squared_error, r2_score
+import joblib
 
-# Load the dataset from Hugging Face
-dataset = load_dataset('Esmaeilkianii/croplogdehkhoda', split='train')
-data = pd.DataFrame(dataset)
+# Load the data
+data = pd.read_csv('sugar_cane_data.csv')
 
-# Split the data into different sections based on columns or logic
-# Assuming the dataset contains the combined information for all four files
+# Data preprocessing
+data['Farm'] = data['Farm'].astype(str)
+data['Variety'] = data['Variety'].fillna('Unknown')
+data['Age'] = data['Age'].fillna('Unknown')
 
-# Separate different sections of the data
-cleaned_data = data[['مزرعه', 'year', 'yield', 'age', 'variety', 'area']]  # Example farm performance data
-veg_index_data = data[['مزرعه', 'date', 'mean_VH', 'mean_VV']]  # Example vegetation index data
-weather_data = data[['سال', 'ماه', 'روز', 'تاریخ', 'درجه حرارت حداقل', 'درجه حرارت حداکثر', 'میانگین درجه حرارت',
-                     'حداقل رطوبت نسبی (درصد)', 'حداکثر رطوبت نسبی (درصد)', 'میانگین رطوبت نسبی (درصد)',
-                     'میزان تبخیر روزانه (میلیمتر)', 'میزان بارندگی (میلیمتر)', 'میزان ساعت افتابی']]  # Weather data
-performance_estimation_data = data[['مزرعه', 'yield', 'tonnage']]  # Example yield and tonnage prediction
+# Split features and target variables
+X = data[['Farm', 'Variety', 'Age']]
+y = data[['Brix', 'Purity', 'Pol', 'RS']]
 
-# Function to search for farm performance
-def search_farm(farm_name):
-    # Filter the farm's data for recent years (1392-1402)
-    farm_data = cleaned_data[cleaned_data['مزرعه'] == farm_name]
-    farm_performance = farm_data[(farm_data['year'] >= 1392) & (farm_data['year'] <= 1402)]
-    
-    # Create a time-series chart for the vegetation index (mean_VH, mean_VV)
-    veg_data = veg_index_data[veg_index_data['مزرعه'] == farm_name]
-    
-    # Plotting vegetation index trends
-    fig, ax = plt.subplots(figsize=(10,5))
-    ax.plot(veg_data['date'], veg_data['mean_VH'], label='mean_VH')
-    ax.plot(veg_data['date'], veg_data['mean_VV'], label='mean_VV')
-    ax.set_xlabel('Date')
-    ax.set_ylabel('Vegetation Index')
-    ax.set_title(f'Vegetation Index Trends for {farm_name}')
-    ax.legend()
-    plt.xticks(rotation=45)
-    
-    return farm_performance, fig
+# Split the data into training and testing sets
+X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
 
-# Function to display weather data
-def show_weather_data():
-    st.dataframe(weather_data)
+# Create preprocessing steps
+preprocessor = ColumnTransformer(
+    transformers=[
+        ('num', StandardScaler(), []),
+        ('cat', OneHotEncoder(handle_unknown='ignore'), ['Farm', 'Variety', 'Age'])
+    ])
 
-# Function to predict farm performance (yield and tonnage)
-def predict_performance(farm_name):
-    prediction_data = performance_estimation_data[performance_estimation_data['مزرعه'] == farm_name]
-    
-    if prediction_data.empty:
-        st.write("مزرعه مورد نظر پیدا نشد.")
-    else:
-        yield_value = prediction_data['yield'].values[0]
-        tonnage = prediction_data['tonnage'].values[0]  # Adjust based on your actual column name
-        st.write(f"تن در هکتار: {yield_value}")
-        st.write(f"تناژ: {tonnage}")
+# Create a pipeline with preprocessor and random forest regressor
+model = Pipeline([
+    ('preprocessor', preprocessor),
+    ('regressor', RandomForestRegressor(n_estimators=100, random_state=42))
+])
 
-# Streamlit app
-st.title("Farm Performance Dashboard")
+# Fit the model
+model.fit(X_train, y_train)
 
-# Sidebar menu
-menu = st.sidebar.selectbox("انتخاب منو", ["Farm Search", "دیتابیس هواشناسی", "پیش بینی عملکرد"])
+# Make predictions on the test set
+y_pred = model.predict(X_test)
 
-# Farm Search section
-if menu == "Farm Search":
-    st.header("Search for Farm Performance")
-    farm_name = st.text_input("نام مزرعه را وارد کنید")
-    
-    if st.button("Search"):
-        farm_performance, veg_chart = search_farm(farm_name)
-        st.write("Farm Performance (1392-1402):")
-        st.dataframe(farm_performance)
-        st.write("Vegetation Index Time Series:")
-        st.pyplot(veg_chart)
+# Evaluate the model
+mse = mean_squared_error(y_test, y_pred)
+r2 = r2_score(y_test, y_pred)
 
-# Weather Database section
-elif menu == "دیتابیس هواشناسی":
-    st.header("Weather Data")
-    st.write("م��اهده داده‌های هواشناسی:")
-    show_weather_data()
+print(f"Mean Squared Error: {mse}")
+print(f"R-squared Score: {r2}")
 
-# Performance Prediction section
-elif menu == "پیش بینی عملکرد":
-    st.header("Predict Farm Performance")
-    farm_name_prediction = st.text_input("نام مزرعه برای پیش بینی")
-    
-    if st.button("پیش بینی"):
-        predict_performance(farm_name_prediction)
+# Save the model
+joblib.dump(model, 'sugar_cane_model.joblib')
 
+# Function to make predictions for new data
+def predict_sugar_cane_properties(farm, variety, age):
+    new_data = pd.DataFrame({'Farm': [farm], 'Variety': [variety], 'Age': [age]})
+    prediction = model.predict(new_data)
+    return {
+        'Brix': prediction[0][0],
+        'Purity': prediction[0][1],
+        'Pol': prediction[0][2],
+        'RS': prediction[0][3]
+    }
+
+# Example usage
+print(predict_sugar_cane_properties('01-18', 'CP69', 'P'))