We are working on the puzzles in the Watermelon-book.
Note that we are not sharing the answers but implementation of the algorithms. For example, class KNeighborsClassifier instead of 'runing knn on iris'. Here is the release package of the algorithms which you can install with pip install cquai-ml.
Detailed puzzle and solution list can be found in the Puzzles.md in each folder. Here is a brief one:
| Algorithm | Description | Code |
|---|---|---|
| Dataset Space | Automatically generate the sample_space, hypothesis_space and version_space for a data set | PY |
| Union Hypothesis Space | Figure out how the number of hypothsis change with the disjunction become longer | CPP PY |
| Linear Regression | Binary Linear regression based on gradient descent (l-bfgs-b) without penalty | PY |
| Logistic Regression | Binary Logistic regression based on gradient descent (l-bfgs-b) without penalty | PY |
| KNN | K-Nearest Neighbors Classifier | PY |
| ... | More puzzles and solutions can be found in the Puzzles.md in each folder. |