First graph is overlayed/duplicated on top of the second graph

🎈 Using Streamlit
1

Summary

I am plotting two graphs with pandas.groupby.plot() and streamlit.pyplot(). The first graph is plotted correctly but the second graph is just the first graph duplicated and on top of it the second graph is plotted (merged into one)

Steps to reproduce

Code snippet:

# First Graph

fig1, ax1 = plt.subplots(figsize=(20,8))
full_count_time_series_plot = df.groupby(
				[df['Time_Series'].dt.year, df['Time_Series'].dt.month]
			).size().plot(kind='line', title='Number of Tickets Over Time', figsize=(20, 8), xlabel='(Year, Month)', ylabel='Count of Tickets', ax=ax1, legend=True, label='Total Tickets')
full_violation_count_time_series_plot = df.groupby(
				[df[df['Time_Taken'] < 0]['Time_Series'].dt.year, df[df['Time_Taken'] < 0]['Time_Series'].dt.month]
			).size().plot(kind='line', title='SLA Violations Over Time', figsize=(20,8), xlabel='(Year, Month)', ylabel='Count of Tickets', ax=ax1, legend=True, label='Violated Tickets')
st.pyplot(fig1)

# Second Graph
st.header('Frequency of Tickets - Type Wise')
type_var = st.selectbox("Select Any Categorical Column", df.columns, index=6)
top_plot = df[type_var].value_counts()[:10].plot(kind='bar', title=f'Number of Tickets in each {type_var}', figsize=(10, 5))
top_fig = top_plot.figure
st.pyplot(top_fig)

If applicable, please provide the steps we should take to reproduce the error or specified behavior.

Expected behavior:
First Graph (Correct)

image
image1811×745 87.3 KB

Second Graph

image
image1167×462 4.73 KB

Actual behavior:
First Graph (Correct)

image
image1809×738 87.3 KB

Second Graph
image
image1169×579 36.6 KB

The second graph has legend from the first graph and the both lines from the first graph (line chart)

Debug info

  • Streamlit version: 1.20.0
  • Python version: 3.9.13
  • Using pip, venv
  • OS version: Alma Linux 8
  • Browser version: Chrome 111.0.5563.65

Requirements file

streamlit
pandas

2

What about creating a separate second figure to put that bar chart in.

bottom_fig, ax = plt.subplots(figsize=(10, 5))
st.header('Frequency of Tickets - Type Wise')
type_var = st.selectbox("Select Any Categorical Column", df.columns, index=6)
bar_chart = df[type_var].value_counts()[:10].plot(
    kind='bar', 
    title=f'Number of Tickets in each {type_var}', ax=ax)
st.pyplot(bottom_fig)
2 Likes
3

Thanks for the solution, the problem is resolved. This is embarrassing but I wasn’t calling subplots() for any of my plots, then trying to fix the issue I added for the first one but forgot to do the same for the second one.