Python Sklearn auc_roc_curve and roc_curve functions don't seem to match

I have the following Python code to compute both AUC and plot the ROC graphic:

import numpy as np
import sklearn.metrics
from sklearn.metrics import roc_curve
from sklearn.metrics import roc_auc_score
from matplotlib import pyplot


testY = np.array([1., 1., 2., 1., 1., 1., 2., 1., 1., 1., 2., 2., 1., 2., 1., 1., 2., 1., 2., 2., 1., 2., 1., 2.,
 1., 2., 1., 1., 1., 1., 1., 2., 2., 2., 1., 1., 1., 1., 1., 1., 2., 2., 1., 1., 1., 1., 1., 2.,
 1., 1., 1., 2., 1., 2., 2., 1., 2., 2., 2., 1., 2., 1., 2., 1., 1., 2., 1., 2., 2., 1., 1., 1.,
 2., 2., 1., 1., 1., 2., 1., 1., 2., 2., 1., 1., 1., 2., 1., 1., 2., 1., 2., 1., 1., 2., 1., 1.,
 1., 2., 1., 2., 1., 2., 2., 1., 2., 1., 1., 1., 1., 1., 2., 2.])

predY = np.array([1.00000000e+00, 0.00000000e+00, 0.00000000e+00, 1.00000000e+00
, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00
, 0.00000000e+00, 0.00000000e+00, 1.00000000e+00, 0.00000000e+00
, 1.00000000e+00, 1.00000000e+00, 0.00000000e+00, 0.00000000e+00
, 1.00000000e+00, 0.00000000e+00, 1.00000000e+00, 1.00000000e+00
, 1.00000000e+00, 0.00000000e+00, 1.00000000e+00, 1.00000000e+00
, 1.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00
, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 1.00000000e+00
, 6.02987735e-03, 0.00000000e+00, 1.00000000e+00, 0.00000000e+00
, 1.00000000e+00, 1.00000000e+00, 0.00000000e+00, 0.00000000e+00
, 0.00000000e+00, 0.00000000e+00, 1.00000000e+00, 0.00000000e+00
, 0.00000000e+00, 1.00000000e+00, 0.00000000e+00, 1.00000000e+00
, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00
, 0.00000000e+00, 1.00000000e+00, 0.00000000e+00, 0.00000000e+00
, 0.00000000e+00, 1.00000000e+00, 1.00000000e+00, 0.00000000e+00
, 1.00000000e+00, 0.00000000e+00, 1.00000000e+00, 0.00000000e+00
, 0.00000000e+00, 1.00000000e+00, 0.00000000e+00, 1.17126142e-37
, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00
, 0.00000000e+00, 3.92363570e-44, 0.00000000e+00, 0.00000000e+00
, 1.00000000e+00, 0.00000000e+00, 1.00000000e+00, 1.00000000e+00
, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 4.22579080e-01
, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00
, 3.12829528e-14, 0.00000000e+00, 1.00000000e+00, 0.00000000e+00
, 0.00000000e+00, 0.00000000e+00, 1.00000000e+00, 1.00000000e+00
, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 1.00000000e+00
, 1.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00
, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00
, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00])

print(testY.shape)
print(predY.shape)
print(predY.min(), predY.max())
lr_auc = roc_auc_score(testY, predY, labels= [1,2]) 

print(lr_auc) 

ns_probs = np.empty( shape = predY.shape[0] )
ns_probs[:] = 0.5

ns_auc = roc_auc_score(testY, ns_probs, labels= [1,2])
print(ns_auc)

ns_fpr, ns_tpr, _ = roc_curve(testY, ns_probs, pos_label = 1) 
lr_fpr, lr_tpr, _ = roc_curve(testY, predY, pos_label = 1)

print("ROC")
print("lr_fpr")
print(lr_fpr)
print("lr_tpr")
print(lr_tpr)

pyplot.plot(ns_fpr, ns_tpr, linestyle='--', label='')
pyplot.plot(lr_fpr, lr_tpr, marker='.', label="Positive label")
# axis labels
pyplot.xlabel('False Positive Rate')
pyplot.ylabel('True Positive Rate')
# show the legend
pyplot.legend()
# show the plot
pyplot.show()

This is the output I get:

(112,)
(112,)
0.0 1.0
0.5855614973262032 # <--- this is the AUC
0.5
ROC
lr_fpr
[0.         0.36363636 0.36363636 0.45454545 1.        ]
lr_tpr
[0.         0.25       0.26470588 0.26470588 1.        ]

And this is the plot I get:

So clearly the plot has less than 0.5 AUC, why there is this difference?

A few notes:

• Yes, the labels are 1 and 2. The positive label is 1. It should still work, I think.
• The array is an actual array of probabilities, between 0 and 1. I'm not sure why it doesn't have a lot of intermediate values, but it should still work because I'm using the same array in both places right?

It appears that the plot is correct and the score is incorrect, assuming your predY is supposed to be the probability of your positive class 1. From the documentation for roc_auc_score:

y_score [...] The probability estimates correspond to the probability of the class with the greater label