Timeline for How max_features parameter works in DecisionTreeClassifier?
Current License: CC BY-SA 4.0
12 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 21, 2018 at 7:30 | vote | accept | James Flash | ||
| S Nov 20, 2018 at 19:53 | history | suggested | tuomastik | CC BY-SA 4.0 |
Reformatted, add better title, add link to sklearn docs
|
| Nov 20, 2018 at 8:44 | comment | added | mapto | @MzdR, the two questions are about different parameters, thus not duplicate. | |
| Nov 20, 2018 at 8:44 | comment | added | mapto | Hi James, it seems that my comment was also inaccurate. @Bashar Haddad's answer sounds more convincing to me. If it convinces you as well, I'd suggest you accept it. | |
| Nov 20, 2018 at 6:17 | review | Suggested edits | |||
| S Nov 20, 2018 at 19:53 | |||||
| Nov 20, 2018 at 6:08 | answer | added | Bashar Haddad | timeline score: 17 | |
| Nov 19, 2018 at 21:20 | review | Close votes | |||
| Nov 20, 2018 at 1:16 | |||||
| Nov 19, 2018 at 21:02 | comment | added | MzdR | Possible duplicate of Is max_depth in scikit the equivalent of pruning in decision trees? | |
| Nov 19, 2018 at 16:37 | comment | added | James Flash | Could you provide any example as an answer, please? Because I could find graphs with only one feature condition at each node. For example, check first ten graphs from google.com/… | |
| Nov 19, 2018 at 16:07 | comment | added | mapto |
In the documentation it is stated: "If int, then consider max_features features at each split". Thus, it it is the maximum number of features used in the condition at each node of the tree. Your example is misleading, because even in the case of max_features=2 your splits are using only one feature in the decisions.
|
|
| Nov 19, 2018 at 13:54 | history | edited | James Flash | CC BY-SA 4.0 |
edited body
|
| Nov 19, 2018 at 13:39 | history | asked | James Flash | CC BY-SA 4.0 |