PA5: Decision tree
- Due May 6, 2022 by 11:59pm
- Points 1
- Submitting a file upload
- File Types zip
See this document Links to an external site.for the PA instructions. Upload your code files and the PDF of your report as a zip.
Rubric
Find Rubric| Criteria | Ratings | ||
|---|---|---|---|
|
the decision tree program compiles and/or runs without error
|
|
||
|
the source code is well organized and commented
|
|
||
|
each source code file includes a header indicating who wrote or contributed to the code
|
|
||
|
code from any external sources, or code that relied on outside influence, includes a citation in a comment just above the corresponding code
|
|
||
|
the code is built off of the code base we built in class
|
|
||
|
there is an implemented, working decision tree program
|
|
||
|
in training mode, the decision tree program supports an optional feature to adjust epsilon (the allowed change in accuracy during pruning)
|
|
||
|
in the decision tree program, the `prune` method is correctly implemented
|
|
||
|
in the decision tree program, the `evaluate` method is correctly implemented
|
|
||
|
in the decision tree program, the `findBestThreshold` method is correctly implemented
|
|
||
|
in prediction mode, the decision tree program supports an option for outputting class label confidence scores (one per distinct label in the training data) instead of label predictions
|
|
||
|
in prediction mode, the decision tree program supports an option for outputting class label confidence scores (one per distinct label in the training data) instead of label predictions; these are listed in a consistent order
|
|
||
|
the decision tree program has an accurate "help" message if the wrong number of required command line arguments are provided
|
|
||
|
the submission includes a README.md file
|
|
||
|
the README.md file describes how to compile (if Java) and run your program
|
|
||
|
the README.md file describes what command line arguments, if any, are available
|
|
||
|
the README.md file includes a section that states who worked on the project
|
|
||
|
the README.md file includes a section that describes the performance of three runs: training on `sms/train.csv` and testing on `sms/dev.csv` with the epsilon values 0, 0.05, and 0.10
|
|
||
|
|
|||
