Daml

🌟 **Feature Release**
- `64416675` - Update clusterer class and documentation

* `Clusterer` detector released

This class assists in exploratory data analysis of unlabeled data by identifying duplicates and outliers. Additional information on usage is available in our documentation.

🌟 **Feature Release**
- `278b4dc1` - Release Linter, Duplicates, ImageStats, ChannelStats and Parity

`Linter`, `Duplicates` detectors and `ImageStats`, `ChannelStats`, and `Parity` metrics are now released. The existing metrics available have also been moved into different modules (`detectors` and `workflows`) that better reflect their functionality.

* `detectors`
* Drift detectors: `DriftCVM`, `DriftKS`, `DriftMMD`, `DriftUncertainty` and supporting classes
* Out-of-distribution detectors: `OOD_AE`, `OOD_AEGMM`, `OOD_LLR`, `OOD_VAE`, `OOD_VAEGMM` and supporting classes
* `Linter`
* `Duplicates`
* `metrics`
* `BER`
* `Divergence`
* `Parity`
* `ImageStats`
* `ChannelStats`
* `UAP`
* `workflows`
* `Sufficiency`

🛠️ **Improvements and Enhancements**
- `58263ac7` - Move niter param to evaluate and calculate and retain curve coefficients in output dictionary

This change enhances the output of the `Sufficiency` metric to provide the coefficients for the learning curve by measure/class when running the metric. These parameters were previously recalculated each call to project and plot. The parameters are provided as a `Dict[str, np.ndarray]` under the `_CURVE_PARAMS_` key in the output dictionary.

🌟 **Feature Release**
- `322fc830` - Add parameter `k` to BER estimator for KNN to enable `k>1` for better consistency with ground truth in certain cases

🛠️ **Improvements and Enhancements**
- `07b12ac2` - Fully integrate outlier detection into DAML

Outlier Detection API has been changed. Additional details are available in our documentation.

🌟 **Feature Release**
- `2ed88a07` - Implement Drift Detection Metrics

This change adds 4 types of Drift Detection metrics which allow for the detection of potential drift in the dataset.

* Kolmogorov-Smirnov
* Cramér-von Mises
* Maximum Mean Discrepancy
* Classifier Uncertainty

The conceptual source is derived from [Failing Loudly: An Empirical Study of Methods for Detecting Dataset Shift](https://arxiv.org/abs/1810.11953) and the implementation is derived from [Alibi-Detect v0.11.4](https://github.com/SeldonIO/alibi-detect/tree/v0.11.4).