alibi
alibi copied to clipboard
SeldonIO
Algorithms for explaining machine learning models
When i tried to use IG for my 3inputs model i got this error : explanation = ig.explain(X_test_sample1['input_type_ids'].numpy().argmax(axis=1), baselines=None, target=None) --------------------------------------------------------------------------- ValueError Traceback (most recent call last) in () 2...
alibi `detect` is considering init based config or something similar, which will be good to apply for `explain` at some point as well. Currently implementation details of alibi explainers are...
The Boston dataset which we use in some examples has an ethical problem and should be replaced. Read more here: https://scikit-learn.org/stable/modules/generated/sklearn.datasets.load_boston.html#sklearn.datasets.load_boston Impacted examples: - `cfproto_housing.ipynb` - `ale_regression_boston.ipynb` The above link...
AnchorText offers support for three masked language model: `DistilbertBaseUncased`, `BertBaseUncased`, `RobertaBase`. All previously enumerate classes inherit the `LanguageModel` class and overwrite two methods. For example, `DistilbertBaseUncased`: ```python class DistilbertBaseUncased(LanguageModel): SUBWORD_PREFIX...
How can the integrated gradient explainer be used for a model with multiple outputs (e.g. a multi-task classification model where each output can have few possible values)? Can you please...
The recent [PR checklist](https://github.com/SeldonIO/alibi/blob/master/CONTRIBUTING.md#pr-checklist) outlines what to check before a PR is considered complete, however, a portion of these are mundane checks that can be automated with existing tools, in...
Part of #453 Explore what's the best way to remove the internal state from the `AnchorText` explainer.
Part of #453 Explore what's the best way to remove the internal state from the `AnchorTabular` explainer.
Currently we don't have guidelines for how to manage the state of an explainer object, especially during `explain` calls. An `explain` method may be very complex internally, calling a bunch...
Both `Explainer` objects and `Explanation` objects (returned by calling `explain`) have a `meta` attribute storing information about how the `Explainer` was configured, including information such as user-configured parameters passed to...