using RDatasets, DataFrames iris = dataset("datasets", "iris") using MLJ # using the MLJ framework using MLJModels # loads the modesl MLJ can use e.g. linear regression, decision tree tree_model = @load DecisionTreeClassifier verbosity=1 y, X = unpack(iris, ==(:Species), !=(:Species)) tree_machine …

In the doc I believe it’s explained that the machine is just a wrapper for the model (a container with just the hyperparameters) and essentially the “fitresults” (whatever is learned during the fitting process that needs to be passed on to the predict or transform method). So in this case if …

[image] tlienart: So in this case if you want to recuperate the actual trained tree , please have a look at tree_machine.fitresult ; IIRC the first element of the tuple is the tree that you want That’s correct. Thanks. print_tree(tree_machine.fitresult[1]) print_tree(fitted_params(tree_mach…

Great, we still have a lot to do in order to clarify how to do the many things that users would want to do with an ML toolbox and having people trying things out and telling us where they find things un-intuitive is very helpful! Generally the tutorials is the good first place to get started; th…

Thanks for commenting on your experience. For the record, you can also get the tree printed by increasing the verbosity level of the fit: julia> X, y = @load_iris ((sepal_length = [5.1, 4.9, 4.7, 4.6, 5.0, 5.4, 4.6, 5.0, 4.4, 4.9 … 6.7, 6.9, 5.8, 6.8, 6.7, 6.7, 6.3, 6.5, 6.2, 5.9], sepal_width = …

I am wondering whether we can plot the learned decision tree model within MLJ like the following cite suggests Visualize a Decision Tree in 4 Ways with Scikit-Learn and Python | MLJAR (via Python version of scikitlearn) instead of simply displaying it as ASCII characters as print_tree does.

Cool! Can the same visualization be applied to trees from DecisionTree.jl?

The plot recipe is unfortunately specific to EvoTrees structure. It would require some adaptations for other tree base models.

I suppose decision trees will be around for a while, possibly longer than EvoTrees.jl and DecisionTree.jl. Some AbstractDescisionTree interface, which would define visualizations, would be pretty awesome.

How to visualise the structure of the decision tree built by MLJ?

Specific Domains Machine Learning

jeremiedb November 16, 2020, 8:05pm 7

Latest EvoTrees.jl version (0.5.2), which is integrated with MLJ, introduces a basic tree visualization:

plot(model, 3) # second argument refers to the ith tree of the model

Topic		Replies	Views
Plotting a decision tree and extracting the split criteria Machine Learning	0	379	February 3, 2021
Visualising a decision tree using GraphViz Visualization package	4	3800	June 27, 2019
Error from MLJ Iris example, no method matching? Machine Learning	4	599	November 11, 2021
Face this run time error General Usage	3	767	December 18, 2019
Failed prediction from BetaML DecisionTreeClassifier Machine Learning question	10	685	September 22, 2022

How to visualise the structure of the decision tree built by MLJ?

Related topics