Support for one-hot encoded features in minimization #87
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: abigailt <[email protected]>
Signed-off-by: abigailt <[email protected]>
Signed-off-by: abigailt <[email protected]>
…ons for 1-hot encoded features are consistent. Signed-off-by: abigailt <[email protected]>
Signed-off-by: abigailt <[email protected]>
Signed-off-by: abigailt <[email protected]>
Signed-off-by: abigailt <[email protected]>
apt/minimization/minimizer.py
Outdated
| self.categorical_features = [] | ||
| if categorical_features: | ||
| self.categorical_features = categorical_features | ||
| self.features_to_minimize = features_to_minimize | ||
| self.feature_slices = feature_slices | ||
| if self.feature_slices: | ||
| self.all_one_hot_features = set([str(feature) for encoded in self.feature_slices for feature in encoded]) |
There was a problem hiding this comment.
You can use set comprehension right away instead of converting list comprehension to set.
| @@ -375,6 +396,8 @@ def fit(self, X: Optional[DATA_PANDAS_NUMPY_TYPE] = None, y: Optional[DATA_PANDA | |||
| x_test_dataset = ArrayDataset(x_test, features_names=self._features) | |||
| self._ncp_scores.fit_score = self.calculate_ncp(x_test_dataset) | |||
| self._ncp_scores.generalizations_score = self.calculate_ncp(x_test_dataset) | |||
| else: | |||
| print('No fitting was performed as some information was missing') | |||
There was a problem hiding this comment.
Can the message be a bit more helpful?
apt/minimization/minimizer.py
Outdated
| elif range['end'] is None and range['start'] > 0: | ||
| feature_value = 1 | ||
| elif range['start'] is not None and range['end'] is not None: | ||
| print(range) |
There was a problem hiding this comment.
What's the feature_value in this case? Seems like an unassigned feature_value will be appended in the next line.
Is the print the only thing that happens here? And even if so, shouldn't some text explain the meaning of this print?
apt/minimization/minimizer.py
Outdated
| feature_value = 1 | ||
| elif range['start'] is not None and range['end'] is not None: | ||
| print(range) | ||
| new_cell['categories'][feature].append(feature_value) |
There was a problem hiding this comment.
And what's the feature_value if none of the ifs match?
apt/minimization/minimizer.py
Outdated
| def _get_other_features_in_encoding(feature, feature_slices): | ||
| for encoded in feature_slices: | ||
| if feature in encoded: | ||
| return (list(set(encoded) - set([feature]))), encoded |
There was a problem hiding this comment.
nit: set([feature]) can be replaced with {feature}
apt/minimization/minimizer.py
Outdated
| new_cell['categories'][other_feature].append(1) | ||
| else: | ||
| new_cell['categories'][other_feature].append(0) | ||
| new_cell['categories'][other_feature].append(1) |
There was a problem hiding this comment.
I'm not sure I understand where this is narrowed down to the single correct value.
…nding so that options are narrowed down Signed-off-by: abigailt <[email protected]>
Signed-off-by: abigailt <[email protected]>
Signed-off-by: abigailt <[email protected]>
Signed-off-by: abigailt <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Any columns in the input data that represent one-hot encoded features can be minimized together to maintain the correctness of the encoding. This is the case both when using the transform method, and in the representative values within the generalizations structure.