How to get precision, recall and f-measure from confusion matrix in Python [duplicate]

Answer 1

There is a package called 'disarray'.

So, if I have four classes :

import numpy as np
a = np.random.randint(0,4,[100])
b = np.random.randint(0,4,[100])

I can use disarray to calculate 13 matrices :

import disarray

# Instantiate the confusion matrix DataFrame with index and columns
cm = confusion_matrix(a,b)
df = pd.DataFrame(cm, index= ['a','b','c','d'], columns=['a','b','c','d'])
df.da.export_metrics()

which gives :