A zoo of labelled datasets and ML models for data prep tasks.
By submitting results, you acknowledge that your holdout test results([X_test.csv,y_test.csv]) are obtained purely by training on the training set([X_train.csv,y_train.csv]).
Model | Overall Accuracy | Usable Numeric Accuracy | Usable-with-Extraction Accuracy | Usable-Categorical Accuracy | Unusable Accuracy | Context-Specific Accuracy |
---|---|---|---|---|---|---|
Random Forest | 89.4 | 95.8 | 79.7 | 93.5 | 74.3 | 83.4 |
k-NN | 88.8 | 95.6 | 77.6 | 87.9 | 79 | 85.6 |
Character-level CNN | 88.3 | 93.6 | 83.9 | 88.2 | 83.2 | 83.2 |
RBF-SVM | 87.4 | 94 | 84.6 | 89.1 | 76 | 80.7 |
Logistic Regression | 86 | 94.5 | 73.4 | 86.8 | 78.4 | 77.2 |