Python结合AI实现数据可视化全流程

import matplotlib.pyplot as plt
import pandas as pd
# 模拟数据：日期和销售额
dates = pd.date_range(start='2023-01-01', periods=12, freq='M')
sales = [100, 120, 90, 150, 180, 200, 220, 190, 210, 230, 250, 280] # 假设的月度销售额
plt.figure(figsize=(10, 6))
plt.plot(dates, sales, marker='o', linestyle='-', color='b')
plt.title('2023 年月度销售趋势')
plt.xlabel('月份')
plt.ylabel('销售额 (万元)')
plt.grid(True)
plt.xticks(rotation=45)
plt.tight_layout()
plt.show()

import seaborn as sns
import pandas as pd
import numpy as np
# 模拟数据：不同产品类别的销售额
np.random.seed(42)
categories = ['电子产品', '服装', '家居', '食品']
data = {
    '类别': np.random.choice(categories, size=100),
    '销售额': np.random.randint(50, 500, size=100) # 随机销售额
}
df = pd.DataFrame(data)
plt.figure(figsize=(8, 6))
sns.boxplot(x='类别', y='销售额', data=df)
plt.title('不同产品类别销售额分布')
plt.xlabel('产品类别')
plt.ylabel('销售额 (元)')
plt.show()

import plotly.express as px
import pandas as pd
# 模拟数据：客户城市、经纬度、客户规模
cities = ['北京', '上海', '广州', '深圳', '杭州', '成都']
latitudes = [39.90, 31.22, 23.12, 22.54, 30.25, 30.67]
longitudes = [116.41, 121.48, 113.25, 114.05, 120.16, 104.06]
sizes = [500, 800, 300, 400, 350, 250] # 客户数量或规模
df = pd.DataFrame({
    '城市': cities,
    '纬度': latitudes,
    '经度': longitudes,
    '规模': sizes
})
fig = px.scatter_geo(df,
                    lat='纬度',
                    lon='经度',
                    size='规模',
                    hover_name='城市',
                    projection="natural earth",
                    title='客户地域分布')
fig.show()

import pandas as pd
import matplotlib.pyplot as plt
import openai  # 需要安装 openai 库并配置 API Key
# 1. 生成一个简单的销售数据图表
months = ['Jan', 'Feb', 'Mar', 'Apr']
sales = [100, 150, 120, 200]
plt.bar(months, sales)
plt.title('Monthly Sales')
plt.ylabel('Sales (k USD)')
plt.savefig('monthly_sales.png')  # 保存图表
# 2. (模拟) 或者，我们直接提供数据描述给 LLM
data_description = f"Sales data for the first four months: January: $100k, February: $150k, March: $120k, April: $200k."
# 3. 调用 OpenAI API 生成分析描述 (需要有效的 API key)
openai.api_key = 'YOUR_OPENAI_API_KEY'  # 替换为你的实际 Key
response = openai.ChatCompletion.create(
    model="gpt-3.5-turbo",
    messages=[
        {"role": "system", "content": "You are a helpful data analyst assistant."},
        {"role": "user", "content": f"Analyze the following sales data and provide a brief summary of key trends and any potential anomalies. Data: {data_description}"}
    ]
)
ai_insight = response.choices[0].message['content']
print("AI Generated Insight:\n", ai_insight)

from fbprophet import Prophet  # 需要安装 fbprophet
import pandas as pd
import matplotlib.pyplot as plt
# 准备数据 (Prophet 需要 ds 和 y 两列)
df = pd.DataFrame({
    'ds': pd.date_range(start='2023-01-01', periods=12, freq='M'),
    'y': [100, 120, 90, 150, 180, 200, 220, 190, 210, 230, 250, 280]
})
# 创建并拟合模型
model = Prophet()
model.fit(df)
# 创建未来时间框架 (预测未来 6 个月)
future = model.make_future_dataframe(periods=6, freq='M')
forecast = model.predict(future)
# 可视化预测结果
fig = model.plot(forecast)
plt.title('销售历史数据与未来预测')
plt.xlabel('日期')
plt.ylabel('销售额 (万元)')
plt.show()
# 可选：可视化预测组件 (趋势、季节性)
fig_components = model.plot_components(forecast)
plt.show()

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.ensemble import IsolationForest
# 生成包含异常点的模拟数据
np.random.seed(42)
x = np.random.randn(100)
y = 2 * x + np.random.randn(100) * 0.5
# 添加几个异常点
x = np.append(x, [3, -2, 4])
y = np.append(y, [10, -5, 0])
df = pd.DataFrame({'Feature1': x, 'Feature2': y})
# 使用 Isolation Forest 检测异常
model_if = IsolationForest(contamination=0.05)  # 假设异常点占比约 5%
model_if.fit(df)
df['anomaly'] = model_if.predict(df)
# 预测值：-1 表示异常，1 表示正常
# 可视化，异常点用红色高亮
plt.figure(figsize=(10, 6))
plt.scatter(df['Feature1'][df['anomaly'] == 1], df['Feature2'][df['anomaly'] == 1], c='blue', label='正常点')
plt.scatter(df['Feature1'][df['anomaly'] == -1], df['Feature2'][df['anomaly'] == -1], c='red', marker='X', s=100, label='异常点')
plt.title('特征散点图（异常点高亮）')
plt.xlabel('Feature 1')
plt.ylabel('Feature 2')
plt.legend()
plt.grid(True)
plt.show()

import streamlit as st
import pandas as pd
import plotly.express as px
# 标题
st.title('智能销售数据分析仪表板')
# 加载数据
@st.cache_data
def load_data():
    # 这里替换为实际加载数据的代码
    return pd.DataFrame({'Month': ['Jan', 'Feb', 'Mar'], 'Sales': [100, 150, 120]})
df = load_data()
# 显示数据
st.subheader('原始数据')
st.dataframe(df)
# 绘制交互式图表
st.subheader('月度销售趋势')
fig = px.line(df, x='Month', y='Sales', title='月度销售')
st.plotly_chart(fig)
# (此处可加入前面提到的AI分析结果展示，例如预测、异常点、文本洞察等)