Results for Exercise Sheet 11 (classification with Perceptron)
Please add your row to the table below, following the examples already there. Specify the precision of your predictions on the test set. Specify in percent, rounded to the next integer. Briefly specify how you decided to stop training and any refinements you might have implemented.
Name |
Genres |
Ratings |
Additional Info |
Elmar |
86% |
71% |
baseline, stop after 10 iterations |
Robin |
86% |
72% |
30 iterations, bigrams |
David |
84% |
71% |
15 iterations |
ES |
84% |
71% |
stop if wrong predictions for trainingset are less than 10 % of documents |
Julian |
85% |
70% |
stop training if a specific percentage of right predictions is hit |
Frank |
86% |
68% |
3 iterations |
Alex |
86% |
71% |
10 iterations, touch each entry one per iteration in random order |
Janosch |
84% |
71% |
20 iterations |
Hui Hui |
80% |
59% |
20 iterations |
Daniel |
85% |
71% |
3 iterations |
Frank2 |
81% |
72% |
10 iterations |
ID |
83% |
71% |
15 iterations |
Divo Silesia |
79% |
69% |
30 iterations |
Matthias |
86% |
65% |
stop when training error improvement < 1% |
Jan |
86% |
67% |
10 iterations |