Recent Releases of dtreeviz
dtreeviz - 2.2.2
See https://github.com/parrt/dtreeviz/milestone/40?closed=1
Contributors
- 2 items: Tudor Lapusan (tlapusan)
- 1 items: StepanWorkV (StepanWorkV)
- 1 items: Tyrion (hxhxd)
- Jupyter Notebook
Published by parrt over 2 years ago
dtreeviz - 2.2.1
Issues fixed
- TypeError: dict() argument after ** must be a mapping, not float (compatibility)
- Visualisation error when a node will contains samples from only one class (sklearn) (bug)
- Pie charts show for data that classifier was trained on and not new data (enhancement)
Improvements, features
- Make sklearn visualisations to support validation datasets. (enhancement)
- Pie charts show for data that classifier was trained on and not new data (enhancement)
- Add different display types for leaf_distributions x-axis. (enhancement)
Contributors
- 4 items: Tudor Lapusan (tlapusan)
- 1 items: Alex Moldovan (Alex-AMC)
- 1 items: Rajdeep (RajdeepTarafder)
- 1 items: HannahAlexander
- 1 items: Matthew Epland (mepland)
- Jupyter Notebook
Published by parrt almost 3 years ago
dtreeviz - 2.1.0
This is a major cleanup release, with @mepland doing most of the heavy lifting; there are refactorings, if you bug fixes, and lots of visualization clean up.
Issues fixed
- TF handle multi class labels for view visualisation (bug)
- Standardize Capitalization (clean up)
- KeyError when using decision_boundaries function (clean up)
- Dimension issue in dtreevizsklearnpipeline_visualisations.ipynb (compatibility)
- font name/size not respected (bug)
- Boston dataset no longer avaliable in sklearn for classifier-decision-boundaries.ipynb and classifier-boundary-animations.ipynb (clean up)
ctree_leaf_distributions()should use the classnames not classes (bug)- rtreeviz_univar() does not show X feature properly (bug)
- class name font could not change (bug)
- labels too close to the axes overlap for classifiers (enhancement)
Improvements, features
- Order explainpredictionpath() (enhancement)
- Improve ctreevizunivar() (enhancement)
- Fix Digraph Graphics (enhancement)
- Allow None legend edge (enhancement)
- ctreeleafdistributions() labelallleaves parameter (enhancement)
- Allow different left and right arrow colors via 'larrow' and 'rarrow' (enhancement)
- Add matplotlib default colors as mpl_colors (enhancement)
- Add option to plot classifier leaves as horizontal bars (enhancement)
- labels too close to the axes overlap for classifiers (enhancement)
Contributors
- 19 items: Matthew Epland (mepland)
- 5 items: Terence Parr (parrt)
- 2 items: Tudor Lapusan (tlapusan)
- 1 items: 0ptimista
- 1 items: Tobias Windisch (windisch)
- 1 items: HidenariUoi
- Jupyter Notebook
Published by parrt about 3 years ago
dtreeviz - 2.0.0
This release re-organizes the API to focus on using a model adaptor that adapts the visualization library to the various supported decision tree libraries.
We simplified the README and rebuilt all of the library-specific notebooks to demonstrate the new API, using a common set of examples:
- sklearn-based examples (colab)
- LightGBM-based examples (colab)
- Spark-based examples (colab)
- TensorFlow-based examples (colab)
- XGBoost-based examples (colab)
- Classifier decision boundaries for any scikit-learn model.ipynb (colab)
- Changing colors notebook (specific to sklearn) (colab)
New API:
``` from sklearn.datasets import load_iris from sklearn.tree import DecisionTreeClassifier import dtreeviz
iris = load_iris() X = iris.data y = iris.target
clf = DecisionTreeClassifier(maxdepth=4) clf.fit(X, y) vizmodel = dtreeviz.model(clf, Xtrain=X, ytrain=y, featurenames=iris.featurenames, targetname='iris', classnames=iris.target_names)
v = viz_model.view() # render as SVG into internal object ```
Previous API: Previously, we did something like this to call functions and pass in the various details of the model and training data:
from dtreeviz.trees import dtreeviz
dtreeviz(tree_model=clf, X_train, ...)
Using old functions with 2.0+:
For backward compatibility to call function dtreeviz() and the old API, you can change the import to be:
from dtreeviz import *
dtreeviz(tree_model=clf, X_train, ...)
Argument name changes:
If you were previously using internal model adaptors, such as ShadowLightGBMTree, please note we have changed the following argument names: x_data->X_train and y_data->y_train.
Stuff we completed:
https://github.com/parrt/dtreeviz/milestone/30?closed=1
- Jupyter Notebook
Published by parrt about 3 years ago
dtreeviz - Fix bar chart viz weirdness
Fixing this bug: https://github.com/parrt/dtreeviz/issues/196
- Jupyter Notebook
Published by parrt about 3 years ago
dtreeviz - 1.3.6
This release makes a small change that really improve the speed of dtreeviz.interpretation.explain_prediction_plain_english().
See Fix: xgbdecisiontree shouldGoLeftAtSplit -- massive-speed-up #182.
- Jupyter Notebook
Published by parrt almost 4 years ago
dtreeviz - 1.3.5
Quick bug fix for scaling images.
- Jupyter Notebook
Published by parrt almost 4 years ago
dtreeviz - 1.3.4
Bug fix release: closed PRs
- Jupyter Notebook
Published by parrt almost 4 years ago
dtreeviz - 1.2 Add LightGBM, sklearn pipelines, bug fixes
See Bugs fixes and pull requests.
The main improvement is the introduction of support for LightGBM, thanks to @tlapusan.
- Jupyter Notebook
Published by parrt almost 5 years ago
dtreeviz - Minor improvements
See issues addressed
- Jupyter Notebook
Published by parrt about 5 years ago
dtreeviz - 1.1.3 Add optional title to dtreeviz plots
- Jupyter Notebook
Published by parrt about 5 years ago
dtreeviz - Fix minor issue with single trees for spark
- Jupyter Notebook
Published by parrt over 5 years ago
dtreeviz - 1.0 Add xgboost support
Most important: Introduces XGBoost support thanks to @tlapusan
Enhancements
- https://github.com/parrt/dtreeviz/pull/91
- https://github.com/parrt/dtreeviz/pull/84
Bugs:
- https://github.com/parrt/dtreeviz/pull/90
- Jupyter Notebook
Published by parrt over 5 years ago
dtreeviz - 0.8.2 add min_samples_leaf to rtreeviz_* functions
- Jupyter Notebook
Published by parrt almost 6 years ago
dtreeviz - 0.8.1 Quick enhancement to allow scale arg for dtreeviz()
Quick enhancement: Add scale arg to dtreeviz()
- Jupyter Notebook
Published by parrt about 6 years ago
dtreeviz - 0.8 New features, bug fixes
Improvements
Fixes:
- improve documentation on coloring doc enhancement
- the right arrows in path to test data are not highlighted bug
- Jupyter Notebook
Published by parrt about 6 years ago
dtreeviz - 0.7.1 Small fixes, improvements
- Add histogram of leaf sizes with option to filter the leaves with less than minsamples or more than maxsamples.
- Added more markersize args
- don't show horizontal bars indicating prediction if no splits requested
- add new color settings:
'scatter_marker_alpha': 0.5, 'tesselation_alpha': 0.3, 'tesselation_alpha_3D' : 0.5,
- Jupyter Notebook
Published by parrt over 6 years ago
dtreeviz - 0.7
Added cool leaf viz features from @tlapusan viz_leaf_samples() and ctreeviz_leaf_samples.
Also, allowed ax=None now in the rtree_* and ctree_* functions.
Fixed:
- indexing wrt class_names & relevant issues
- added type check and conversion for y_train parameter
- the right arrows in path to test data are not highlighted
- Jupyter Notebook
Published by parrt over 6 years ago
dtreeviz - Configurable colors, add colorbar, code cleanup
See changes for 0.5. Thanks to @Ashafix and @H4dr1en for the PRs:
- Configurable colors
- Added colorbar
- code clean up
- Jupyter Notebook
Published by parrt over 6 years ago
dtreeviz - 0.4
Lots of bug fixes but main improvement is drastically simplified install procedure for OS X and Windows, thanks to @gautamkarnik. Also, I merged @h4nyu's PR for remove hard coded fontname.
- Jupyter Notebook
Published by parrt over 6 years ago
dtreeviz - Add uni and bivariate feature space depictions
Regression univariate feature-target space
Regression bivariate feature-target space
Regression bivariate feature-target space heatmap
Classification univariate feature-target space
Classification bivariate feature-target space
- Jupyter Notebook
Published by parrt about 7 years ago
dtreeviz - Use simple node names to avoid issues with weird feature names
- Jupyter Notebook
Published by parrt over 7 years ago
dtreeviz - Fix windows compatibility issue
Just a tweak to fix path issue for windows. thanks to @whamp.
- Jupyter Notebook
Published by parrt over 7 years ago
dtreeviz - make dtreeviz cross-platform
Enhancements:
- Squashed all commits prior to these updates to remove huge legacy files
- Works across mac, linux, windows 10. Only .svg output on linux, windows
- Leaf nodes respect class weights now in classifier trees, thanks to @dschoenleber
- SVG files on all platforms embed images for nodes, rather than linking to them. We had to do this with custom XML munging code, except on OS X where we can do
-Tsvg:cairoondotcommand - Legend on classifiers now done with mathplotlib not graphviz
- Jupyter Notebook
Published by parrt over 7 years ago
dtreeviz - Initial (re-)release with decision tree visualization
I squashed commit history due to huge set of dead files. renamed from animl to detreeviz
- Jupyter Notebook
Published by parrt over 7 years ago