I am using the following nested dictionary to make a lineplot:
df = {'A':
{'weight': [200, 190, 188, 180, 170],
'days_since_gym': [0, 91, 174, 205, 279],
'days_since_fasting': 40},
'B':
{'weight': [181, 175, 172, 165, 150],
'days_since_gym': [43, 171, 241, 273, 300],
'days_since_fasting': 100}}
While making the lineplot, I want the Y-Axis
ticks as the percentage
value, for which I'm using PercentFormatter
:
# set the plot size
fig, ax = plt.subplots(2, figsize=(10, 6))
for i, x in enumerate(df.keys()):
sns.lineplot(
x=df[x]['days_since_gym'],
y=df[x]['weight'],
marker="o",
ax=ax[i],
)
ax[i].axvline(df[x]['days_since_fasting'], color='k', linestyle='--', label='Fasting Starts')
ax[i].set_xlim(left=0, right=365)
# Percentage y-axis
ax[i].yaxis.set_major_formatter(mtick.PercentFormatter())
plt.xlabel('Days Since Joined Gym')
plt.ylabel('Relastive Weight')
plt.legend(bbox_to_anchor=(1.04, 1), loc="upper left")
plt.show()
However, I don't want the default percentage value (as the fig shows). I want the 1st value
would be the starting percentage
and the subsequent value would be the relative percentage
. For example, the 1st plot starts with 200%
, which I want as 0%
, and the plot ends with 170%,
which I want as -something%
.
Any suggestions would be appreciated. Thanks!
CodePudding user response:
One way with minor changes to your code is to make the values in y
relative to the first value. That is, keep everything as is and replace:
y=df[x]['weight'],
with:
y=[a-df[x]['weight'][0] for a in df[x]['weight']],