app.py

import streamlit as st
import pandas as pd
import matplotlib.pyplot as plt
import io
import numpy as np
from sklearn.linear_model import LinearRegression
from sklearn.preprocessing import PolynomialFeatures
from sklearn.pipeline import make_pipeline
from sklearn.svm import SVR
from sklearn.ensemble import RandomForestRegressor

st.title("Webcam Color Detection Charting")

uploaded_file = st.file_uploader("Choose a CSV file", type="csv")

time_frame_options = [
    "All",
    "1 second",
    "5 seconds",
    "10 seconds",
    "30 seconds",
    "1 minute",
    "5 minutes",
    "10 minutes",
    "30 minutes",
    "60 minutes",
]
time_frame = st.selectbox("Data Time Frame", time_frame_options)


regression_options = [
    "None",
    "Linear Regression",
    "Polynomial Regression",
    "SVR (Support Vector Regression)",
    "Random Forest Regression",
]
regression_type = st.selectbox("Regression Analysis Type", regression_options)
if uploaded_file is not None:
    # CSV 파일 읽기
    data = pd.read_csv(uploaded_file)

    # 시간 프레임에 따른 데이터 필터링
    if time_frame != "All":
        seconds = {
            "1 second": 1,
            "5 seconds": 5,
            "10 seconds": 10,
            "30 seconds": 30,
            "1 minute": 60,
            "5 minutes": 300,
            "10 minutes": 600,
            "30 minutes": 1800,
            "60 minutes": 3600,
        }
        data['timestamp'] = pd.to_datetime(data['timestamp'], unit='ms')
        data.set_index('timestamp', inplace=True)
        data = data.resample(f"{seconds[time_frame]}S").mean().dropna().reset_index()

    # 차트 생성
    fig, axes = plt.subplots(2, 1, figsize=(10, 8))

    # RGB 차트
    axes[0].plot(data['R'], 'r', label='R')
    axes[0].plot(data['G'], 'g', label='G')
    axes[0].plot(data['B'], 'b', label='B')
    
# 회귀 분석 수행
X = np.arange(len(data)).reshape(-1, 1)

# 선형 회귀
if regression_type == "Linear Regression":
    model = LinearRegression()
    model.fit(X, data['R'])
    axes[0].plot(X, model.predict(X), 'r--')
    st.write(f"R: y = {model.coef_[0]} * x + {model.intercept_}")
    model.fit(X, data['G'])
    axes[0].plot(X, model.predict(X), 'g--')
    st.write(f"G: y = {model.coef_[0]} * x + {model.intercept_}")
    model.fit(X, data['B'])
    axes[0].plot(X, model.predict(X), 'b--')
    st.write(f"B: y = {model.coef_[0]} * x + {model.intercept_}")

# 다항 회귀
elif regression_type == "Polynomial Regression":
    polynomial_features = PolynomialFeatures(degree=2)
    model = make_pipeline(polynomial_features, LinearRegression())
    model.fit(X, data['R'])
    axes[0].plot(X, model.predict(X), 'r--')
    model.fit(X, data['G'])
    axes[0].plot(X, model.predict(X), 'g--')
    model.fit(X, data['B'])
    axes[0].plot(X, model.predict(X), 'b--')
    st.write("Polynomial regression equation is not easily representable.")

# SVR (Support Vector Regression)
elif regression_type == "SVR (Support Vector Regression)":
    model = SVR()
    model.fit(X, data['R'])
    axes[0].plot(X, model.predict(X), 'r--')
    model.fit(X, data['G'])
    axes[0].plot(X, model.predict(X), 'g--')
    model.fit(X, data['B'])
    axes[0].plot(X, model.predict(X), 'b--')
    st.write("SVR equation is not easily representable.")

# Random Forest Regression
elif regression_type == "Random Forest Regression":
    model = RandomForestRegressor()
    model.fit(X, data['R'])
    axes[0].plot(X, model.predict(X), 'r--')
    model.fit(X, data['G'])
    axes[0].plot(X, model.predict(X), 'g--')
    model.fit(X, data['B'])
    axes[0].plot(X, model.predict(X), 'b--')
    st.write("Random Forest equation is not easily representable.")
axes[0].legend(loc='upper right')
    axes[0].set_title('RGB Values')

    # HSV 차트
    axes[1].plot(data['H'], 'r', label='H')
    axes[1].plot(data['S'], 'g', label='S')
    axes[1].plot(data['V'], 'b', label='V')
    axes[1].legend(loc='upper right')
    axes[1].set_title('HSV Values')

    st.pyplot(fig)