decision-tree

问题 Is it possible to extract the node assignment for a fitted rpart tree? What about when I apply the model to new data? The idea is that I would like to use the nodes as a way to cluster my data. In other packages (e.g. SPSS), I can save the predicted class, probabilities, and node number for further analysis. Given how powerful R can be, I imagine there is a simple solution to this. 回答1: Try using the partykit package: library(rpart) z.auto <- rpart(Mileage ~ Weight, car.test.frame) library

问题 I have the following plot as you can see in the picture, Is there any way to see exact number of percentage in the leaf nodes? 回答1: If you want to "see" the percentages, the easiest way is to make a table() of the terminal nodes vs. the response and then look at the conditional proportions. If you want to "see" the proportions in the barplot, then there was no possibility to do this up to now. However, I tweaked the node_barplot() function to accomodate this feature. So if you re-install the

问题 I have trained a decision tree (Python dictionary) as below. Now I am trying to plot it using pydot. In defining each node of the tree (pydot graph), I appoint it a unique (and verbose) name and a brief label. My problem is that in the resulting figure that I get by writing to a .png, I see the verbose node names and not the node labels . I have followed the answer by @Martijn Pieters here. I do not know what I am missing, any ideas? import pydot tree= {'salary': {'41k-45k': 'junior', '46k

问题 What's the best way to handle missing feature attribute values with Weka's C4.5 (J48) decision tree? The problem of missing values occurs during both training and classification. If values are missing from training instances, am I correct in assuming that I place a '?' value for the feature? Suppose that I am able to successfully build the decision tree and then create my own tree code in C++ or Java from Weka's tree structure. During classification time, if I am trying to classify a new

问题 How do I combine few weak learners into a strong classifier? I know the formula, but the problem is that in every paper about AdaBoost that I've read there are only formulas without any example. I mean - I got weak learners and their weights, so I can do what the formula tells me to do (multiply learner by its weight and add another one multiplied by its weight and another one etc.) but how exactly do I do that? My weak learners are decision stumps. They got attribute and treshold, so what do

问题 I am attempting to train models with GradientBoostingClassifier using categorical variables. The following is a primitive code sample, just for trying to input categorical variables into GradientBoostingClassifier . from sklearn import datasets from sklearn.ensemble import GradientBoostingClassifier import pandas iris = datasets.load_iris() # Use only data for 2 classes. X = iris.data[(iris.target==0) | (iris.target==1)] Y = iris.target[(iris.target==0) | (iris.target==1)] # Class 0 has

问题 I would like to inspect all the observations that reached some node in an rpart decision tree. For example, in the following code: fit <- rpart(Kyphosis ~ Age + Start, data = kyphosis) fit n= 81 node), split, n, loss, yval, (yprob) * denotes terminal node 1) root 81 17 absent (0.79012346 0.20987654) 2) Start>=8.5 62 6 absent (0.90322581 0.09677419) 4) Start>=14.5 29 0 absent (1.00000000 0.00000000) * 5) Start< 14.5 33 6 absent (0.81818182 0.18181818) 10) Age< 55 12 0 absent (1.00000000 0

问题 I am training a decision tree with sklearn. When I use: dt_clf = tree.DecisionTreeClassifier() the max_depth parameter defaults to None . According to the documentation, if max_depth is None , then nodes are expanded until all leaves are pure or until all leaves contain less than min_samples_split samples. After fitting my model, how do I find out what max_depth actually is? The get_params() function doesn't help. After fitting, get_params() it still says None . How can I get the actual

问题 I tried to add my C50 tree plot to ppt from R. (ps: not R markdown) However, I am not sure if its because the plot size was too large or not, everytime I ran the code as below, it crashed my R studio. I even tried the code without re-sizing the plot (where I put # to not include that part of code) but it did not work. I tried to add the plot to PDF and it worked and it was clearly displayed. But I really need to figure out a way to put it into the PPTX. Sorry I cannot show you guys the tree

问题 I'm using DecisionTreeClassifier from scikit-learn to classify some multiclass data. I found many posts describing how to display the decision tree path, like here, here, and here. However, all of them describe how to display the tree for the trained data. It makes sense, because export_graphviz only requires a fitted model. My question is how do I visualize the tree on the test samples (preferably by export_graphviz ). I.e. after fitting the model with clf.fit(X[train], y[train]) , and then