Streamlit dashboard 만들기
DB CAFE
thumb_up 추천메뉴 바로가기
- DBA { Oracle DBA 명령어 > DBA 초급 과정 > DBA 고급 과정 }
- 튜닝 { 오라클 튜닝 목록 }
- 모델링 { 데이터 모델링 가이드 }
1 Streamlit dashboard 만들기[편집]
1.1 DashBoard_Plotly.py[편집]
import streamlit as st
import plotly.express as px
import pandas as pd
import os
import warnings
warnings.filterwarnings('ignore')
st.set_page_config(page_title="Superstore!!!", page_icon=":bar_chart:",layout="wide")
st.title(" :bar_chart: Sample SuperStore EDA")
st.markdown('<style>div.block-container{padding-top:1rem;}</style>',unsafe_allow_html=True)
fl = st.file_uploader(":file_folder: Upload a file",type=(["csv","txt","xlsx","xls"]))
if fl is not None:
filename = fl.name
st.write(filename)
# df = pd.read_csv(filename,encoding="ISO-8859-1")
df = pd.read_excel(filename)
else:
os.chdir(r"C:\DEV\python")
df= pd.read_excel("Superstore.xls")
col1,col2 = st.columns((2))
df["Order Date"] = pd.to_datetime(df["Order Date"])
# min/max date
startDate = pd.to_datetime(df["Order Date"]).min()
endDate = pd.to_datetime(df["Order Date"]).max()
with col1:
date1 = pd.to_datetime(st.date_input("Start Date",startDate))
with col2:
date2 = pd.to_datetime(st.date_input("End Date",endDate))
df = df[(df["Order Date"] >= date1) & (df["Order Date"] <= date2)].copy()
st.sidebar.header("Choose Filter:")
#region create
region = st.sidebar.multiselect("+ Region",df["Region"].unique())
if not region:
df2 = df.copy()
else:
df2 = df[df["Region"].isin(region)] # 선택한 지역
# state create
state = st.sidebar.multiselect("+ State",df2["State"].unique())
if not state:
df3 = df2.copy()
else:
df3 = df2[df2["State"].isin(state)] # 선택한 주
# city create
city = st.sidebar.multiselect("+ City",df3["City"].unique())
# filter data Region,State,City
if not region and not state and not city: # 모두 비워져 있으면
filtered_df = df
elif not state and not city: # 주, 도시 가 비워져 있으면
filtered_df = df[df["Region"].isin(region)] # 선택된 지역 필터 하고
elif not region and not city: # region,city가 비워져 있으면
filtered_df = df[df["State"].isin(region)]
elif state and city : #
filtered_df = df3[df["State"].isin(state) & df3["City"].isin(city)]
elif region and city : #
filtered_df = df3[df["Region"].isin(region) & df3["City"].isin(city)]
elif region and state : #
filtered_df = df3[df["Region"].isin(region) & df3["State"].isin(state)]
elif city:
filtered_df = df3[df3["City"].isin(city)]
else:
filtered_df = df3[df3["Region"].isin(region) & df3["State"].isin(state) & df3["City"].isin(city)]
# 카테고리 데이터 프레임 합계
category_df = filtered_df.groupby(by = ["Category"], as_index = False )["Sales"].sum()
with col1:
st.subheader("Category wise Sales")
fig = px.bar(category_df
, x = "Category"
, y = "Sales"
, text = ['${:,.2f}'.format(x) for x in category_df["Sales"]]
, template = "seaborn")
st.plotly_chart(fig,use_container_width=True, height = 200)
with col2:
st.subheader("Region wise Sales")
fig = px.pie(filtered_df, values = "Sales", names = "Region", hole = 0.5)
fig.update_traces(text = filtered_df["Region"], textposition = "outside")
st.plotly_chart(fig,use_container_width=True)
############### 데이터 뷰 ###############################
cl1, cl2 = st.columns((2))
with cl1:
with st.expander("Category_ViewData"):
st.write(category_df.style.background_gradient(cmap="Blues"))
csv = category_df.to_csv(index = False).encode('utf-8')
st.download_button("Download Data"
, data = csv
, file_name = "Category.csv"
, mime = "text/csv"
, help = 'Click here to download the data as a CSV file')
with cl2:
with st.expander("Region_ViewData"):
region = filtered_df.groupby(by = "Region", as_index = False)["Sales"].sum()
st.write(region.style.background_gradient(cmap="Oranges"))
csv = region.to_csv(index = False).encode('utf-8')
st.download_button("Download Data"
, data = csv
, file_name = "Region.csv"
, mime = "text/csv"
, help = 'Click here to download the data as a CSV file')
##################
filtered_df["month_year"] = filtered_df["Order Date"].dt.to_period("M")
st.subheader('Time Series Analysis')
linechart = pd.DataFrame(filtered_df.groupby(filtered_df["month_year"].dt.strftime("%Y : %b"))["Sales"].sum()).reset_index()
fig2 = px.line(linechart
, x = "month_year"
, y="Sales"
, labels = {"Sales": "Amount"}
, height=500, width = 1000
, template="gridon")
st.plotly_chart(fig2,use_container_width=True)
with st.expander("View Data of TimeSeries:"):
st.write(linechart.T.style.background_gradient(cmap="Blues"))
csv = linechart.to_csv(index=False).encode("utf-8")
st.download_button('Download Data', data = csv, file_name = "TimeSeries.csv", mime ='text/csv')
# Create a treem based on Region, category, sub-Category
st.subheader("Hierarchical view of Sales using TreeMap")
fig3 = px.treemap(filtered_df
, path = ["Region","Category","Sub-Category"]
, values = "Sales"
, hover_data = ["Sales"]
, color = "Sub-Category")
fig3.update_layout(width = 800, height = 650)
st.plotly_chart(fig3, use_container_width=True)
chart1, chart2 = st.columns((2))
with chart1:
st.subheader('Segment wise Sales')
fig = px.pie(filtered_df, values = "Sales", names = "Segment", template = "plotly_dark")
fig.update_traces(text = filtered_df["Segment"], textposition = "inside")
st.plotly_chart(fig,use_container_width=True)
with chart2:
st.subheader('Category wise Sales')
fig = px.pie(filtered_df, values = "Sales", names = "Category", template = "gridon")
fig.update_traces(text = filtered_df["Category"], textposition = "inside")
st.plotly_chart(fig,use_container_width=True)
import plotly.figure_factory as ff
st.subheader(":point_right: Month wise Sub-Category Sales Summary")
with st.expander("Summary_Table"):
df_sample = df[0:5][["Region","State","City","Category","Sales","Profit","Quantity"]]
fig = ff.create_table(df_sample, colorscale = "Cividis")
st.plotly_chart(fig, use_container_width=True)
st.markdown("Month wise sub-Category Table")
filtered_df["month"] = filtered_df["Order Date"].dt.month_name()
sub_category_Year = pd.pivot_table(data = filtered_df, values = "Sales", index = ["Sub-Category"],columns = "month")
st.write(sub_category_Year.style.background_gradient(cmap="Blues"))
# Create a scatter plot
data1 = px.scatter(filtered_df, x = "Sales", y = "Profit", size = "Quantity")
data1['layout'].update(title="Relationship between Sales and Profits using Scatter Plot."
, titlefont = dict(size=30)
, xaxis = dict(title = "Sales" , titlefont = dict(size=22))
, yaxis = dict(title = "Profit" , titlefont = dict(size=22)))
st.plotly_chart(data1,use_container_width=True)
with st.expander("View Data"):
st.write(filtered_df.iloc[:500,1:20:2].style.background_gradient(cmap="Oranges"))
# Download orginal DataSet
csv = df.to_csv(index = False).encode('utf-8')
st.download_button('Download Data', data = csv, file_name = "Data.csv",mime = "text/csv")