Why seaborn refuses to produce graph? (wide-form data error) [duplicate]

by

I have the following DF.

                        symbol             bid1Price    ask1Price
datetime            
2023-09-13 05:30:05.447 BTC-29SEP23-16000-C 0   0
2023-09-13 05:30:06.600 BTC-29SEP23-17000-C 0   0
2023-09-13 05:30:07.393 BTC-29SEP23-18000-C 5   0
2023-09-13 05:30:08.180 BTC-29SEP23-19000-C 0   0
2023-09-13 05:30:08.629 BTC-29SEP23-20000-C 5   0
...

‘symbol’ is Categorical:

Data columns (total 3 columns):
 #   Column     Non-Null Count  Dtype   
---  ------     --------------  -----   
 0   symbol     2332 non-null   category
 1   bid1Price  2332 non-null   int64   
 2   ask1Price  2332 non-null   int64

The idea is to plot all symbols (aka strikes) at the same graph. If i do .groupby(‘symbol’), then i can plot any strike from groupping. But why SNS fails with error?

sns.relplot(kind='line',
            data=tmp1,
            hue="symbol",
            height=8,
            aspect=2
)
ValueError: The following variable cannot be assigned with wide-form data: `hue`

I cannot undestand the meaning if the error. According to SNS documentaion

hue – vector or key in data
Grouping variable that will produce elements with different colors. Can be either categorical or numeric,

  • 2

    dfm = df.melt(id_vars='symbol', ignore_index=False).reset_index() and sns.relplot(kind='line', data=dfm, x='datetime', y='value', hue='symbol', style='variable', height=8, aspect=2))

    – 




Leave a Comment