Errors with displaying plotly graph objects

šŸ“Š Plotly Python
1

I had been running a code in Jupyter Notebook that had been working fine before. Iā€™m using version 4.5.0 of plotly. The code added traces and output a plot, but now I see the error:

AttributeError: ā€˜Figureā€™ object has no attribute ā€˜update_layoutā€™

I commented out the update_layout part, but now it looks like the figure simply doesnt show up. This is what my original function that was working looks like:

import pandas as pd
import numpy as np
import plotly.graph_objects as go
import datetime as dt


def create_scatter(df,ticker_list,y_column_list,title_name='',xlabel='',ylabel=''):
    fig = go.Figure()
    for y in y_column_list:
        for t in ticker_list:
            fig.add_trace(go.Scatter(x=df[df.entity_code == t].date,y=df[df.entity_code == t][y],name=t))
    fig.update_layout(
                title={'text': title_name,
                               'y':0.9,
                               'x':0.5,
                       'xanchor': 'center',
                       'yanchor': 'top'},
                xaxis_title=xlabel,
                yaxis_title=ylabel,
                showlegend=True)
    fig.show()

Any help would be appreciated, thank you!

Update: Even when I run the example code in the documentation, I get the same attribute error.

import plotly.graph_objects as go

fig = go.Figure()

fig.add_trace(go.Scatter(
x=[0, 1, 2, 3, 4, 5, 6, 7, 8],
y=[0, 1, 2, 3, 4, 5, 6, 7, 8],
name=ā€œName of Trace 1ā€ # this sets its legend entry
))

fig.add_trace(go.Scatter(
x=[0, 1, 2, 3, 4, 5, 6, 7, 8],
y=[1, 0, 3, 2, 5, 4, 7, 6, 8],
name=ā€œName of Trace 2ā€
))

fig.update_layout(
title=ā€œPlot Titleā€,
xaxis_title=ā€œx Axis Titleā€,
yaxis_title=ā€œy Axis Titleā€,
font=dict(
family=ā€œCourier New, monospaceā€,
size=18,
color="#7f7f7f"
)
)

fig.show()

2

Hi @meaydemi! Thanks for debugging your problem up to the point that you could provide a reproducible example, this really helps. So this is not normal at all. Can you confirm that youā€™re using plotly 4.5.0 and not another version by printing within the same notebook session print(plotly.__version__). Thanks!

3

Thanks for the reply Emmanuelle. I realized what my mistake was. I initially had created a new environment and deliberately installed a much older version of Plotly (3.10.0). When I upgraded to 4.5.0 I never actually restarted the kernel in notebook. It seems that even though package installations are read in instantly (for example I also installed sklearn and didnā€™t need to restart kernel), it looks like not everything gets picked up? Anyways, after I did the kernel restart it worked as expected.

4

Cool, happy to know you solved your problem. Yeah, you have to reload(plotly) if you changed the version after you had already installed it. But restarting the kernel is a cleaner solution.