Compare the model performance with trivial models - Githubissues

bzamecnik / ml

Machine learning projects, often on audio datasets

MIT License

92 stars 39 forks source link

Compare the model performance with trivial models #9

Open bzamecnik opened 8 years ago

bzamecnik commented 8 years ago

random guess
majority class

Figure out minimum reasonable metric values: accuracy, hamming score, AUC.