Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

LabelAccuracyIssue Count
Performance100%87
Test99.59%

245

Question97.02%302
Doc90.32%155
Installation84.07%113
Example80.81%99
Bug78.66%389
Build69.87%156
onnx69.57%23
scala67.24%58
gluon44.38%160
flaky42.78%194
Feature32.24%335
C++29.33%75
ci28.30%53
Cuda22.09%86


Language Detection from Code Snippets in Issues: 

LanguageAccuracy
Clojure98.90%
Java98.50%
Python98.30%
C++97.20%
Scala96.30%


*** In depth analysis with precision, recall, and f1 ***

...

LabelPrecisionRecallF1 ScoreCount
Performance100%100%100%87
Test99.59%100%99.8%245
Question100%97.02%98.49%302
Doc100%90.32%94.92%155
Installation100%84.07%91.35%113
Example100%80.81%89.39%99
Bug100%78.66%88.06%389
Build100%69.87%82.26%156
onnx80%84.21%82.05%23
scala86.67%75%80.41%58
gluon62.28%60.68%61.47%160
flaky96.51%43.46%59.93%194
Feature32.43%98.18%48.76%335
C++55%38.6%45.36%75
ci48.39%40.54%44.12%53
Cuda22.09%100%36.19%86

Classification Report for Language Detection from Code Snippets:

LanguagePrecisionRecallF1 Score
Clojure98.31%98.90%98.61%
Python98.70%98.30%98.50%
Java97.24%98.50%97.87%
C++98.28%97.20%97.74%
Scala97.37%96.30%96.84%



Precision here representing how accurate our classifier was in correctly labelling an issue given all the times it had predicted that label.

...