from sklearn.metrics import f1_score as sklearn_f1
from autotagger.evaluation import basic
predicted,actual = basic.build_dummy_pred_test_pair()
# manually calculating f1 score and averaging the results for each instance
basic.get_samples_f1(predicted,actual)
0.7000000000000001
sklearn_f1(actual,predicted,average='samples')
0.70000000000000007
basic.get_micro_f1(predicted,actual)
0.7142857142857143
sklearn_f1(actual,predicted,average='micro')
0.7142857142857143