Upload 2 files

app.py

@@ -0,0 +1,67 @@
+import streamlit as st
+import pandas as pd
+from sklearn.preprocessing import OneHotEncoder
+from sklearn.preprocessing import LabelEncoder
+from sklearn.preprocessing import OneHotEncoder, MinMaxScaler
+import numpy as np
+import pickle
+
+st.header("Predict Hanoi Weather")
+lst_wind_d = ['N', 'NNE', 'NE', 'ENE', 'E', 'ESE', 'SE', 'SSE', 'S', 'SSW', 'SW', 'WSW', 'W', 'WNW', 'NW', 'NNW']
+wind_directions = [
+    "North", "North-Northeast", "Northeast", "East-Northeast",
+    "East", "East-Southeast", "Southeast", "South-Southeast",
+    "South", "South-Southwest", "Southwest", "West-Southwest",
+    "West", "West-Northwest", "Northwest", "North-Northwest"]
+def choose(choose):
+    if choose in wind_directions:
+        index = wind_directions.index(choose)
+        return lst_wind_d[index]
+    else:
+        return None
+
+TPmin = st.slider("Minimum Temperatures(Celsius)", 0, 100, 20)
+Tpmax = st.slider("Maximum Temperatures(Celsius)", 0, 100, 30)
+Wd = st.slider("Wind Speed(km/h)", 0, 150, 10)
+wind_d = st.selectbox("Wind Direction", wind_directions)
+Humidity = st.slider("Humidity(%)", 0, 100, 80)
+Cloud = st.slider("Cloud(%)", 0, 100, 10)
+pressure=st.slider("Pressure(Bar)", 900, 1050, 1000)
+
+data_list = [[Tpmax, TPmin, Wd, wind_d, Humidity, Cloud, pressure]]
+
+# Tạo DataFrame từ danh sách dữ liệu
+df = pd.DataFrame(data_list, columns=["max", "min", "wind", "wind_d", "humidi", "cloud", "pressure"])
+df["wind_d"] = df["wind_d"].astype('category')
+
+def encode_data(encoders, df, feature):
+    encoder = LabelEncoder()
+    df[feature] = encoder.fit_transform(df[feature])
+    encoders[feature] = encoder
+    return encoders, df
+
+numeric_features = ['max', 'min', 'wind', 'humidi', 'cloud', 'pressure']
+categorical_features = ['wind_d']
+label = 'rain'
+
+scalers = {}
+encoders = {}
+
+df_processed = df.copy(deep=True)
+for feature in categorical_features:
+    encoders, df_processed = encode_data(encoders, df_processed, feature)
+
+with open('knn_model.pkl', 'rb') as model_file:
+    knn_model = pickle.load(model_file)
+
+if Tpmax > TPmin:
+    btn = st.button("Predict")
+    if btn:
+        # Thực hiện dự đoán
+        predicted_class = knn_model.predict(df_processed)
+        if predicted_class[0] == 1:
+            st.success("Today, Hanoi has rain.")
+        else:
+            st.success("Today, Hanoi didn't have rain.")
+else:
+    st.error("Maximum Temperature (TPmax) must be greater than Minimum Temperature (TPmin). Please adjust your inputs.")

knn_model.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a6dac4eea540d72bc948a8da5f3b67e0c0ad7f2d34360617734f6f8cccf5362
+size 262704