Computes \(\widehat Y = X \, C \, A_{\mathrm{new}}\)
(\(= X (C_{+} - C_{-})(A_{+}^{\mathrm{new}} - A_{-}^{\mathrm{new}})\)).
For type = "response" the raw prediction is returned
(possibly signed). For type = "prob" and "class",
negative entries of \(\widehat Y\) are clipped to zero before
column normalization, since probabilities must be non-negative.
Usage
# S3 method for class 'nmfkc.signed'
predict(object, newA = NULL, type = c("response", "prob", "class"), ...)
Arguments
- object
A fitted "nmfkc.signed" object.
- newA
Real-valued \(D \times N_{\mathrm{new}}\) covariate matrix.
- type
Output: "response" (raw signed), "prob",
or "class".
- ...
Unused.
Value
A numeric matrix ("response" or "prob") or a
character vector ("class").
Lifecycle
This function is experimental. The interface may change in
future versions.
References
Ding, C. H. Q., Li, T., & Jordan, M. I. (2010). Convex and
semi-nonnegative matrix factorizations. IEEE Transactions on
Pattern Analysis and Machine Intelligence, 32(1), 45–55.