Calculates the exact shapley value for every player. If you have N features/ players, there must be 2ˆN-1 rows in data.input. You can leave the zero coalation away. Every row must be unique. Below you can see an examle data input: test.data = as.data.frame(rbind(c(1, 0, 5), c(0, 1, 4),c(1, 1, 10))) name(test.data) = c("A", "B", "value") And here with zero coalation: test.data = as.data.frame(rbind(c(0, 0, 0), c(1, 0, 5), c(0, 1, 4),c(1, 1, 10))) Player A has a coalation value of 5. Player A and B have a coalation value of 10.
1 | shapley.unsampled(data.input = test, target = "value")
|
data.input |
A table with all possible observations. |
target |
Put there the name of column with the results of the coalation-function. |
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.