You can basically export most of the Neural Network models you build using the R nnet package into PMML by using the PMML package is available from CRAN. With this package, a PMML representation can be obtained for Neural Networks implementing:

- multi-class classification
- binary classifcation
- regression

- Scaling of input variables: Since
does not automatically implement scaling of numerical inputs, you will need to add scaling via the**nnet**package pass that computation to the**pmmlTransformations**package together with the**pmml**object. You will need to do that if you are planning to use the model to compute scores/results from raw data.**nnet** - The PMML exporter uses transformations to create dummy variables for categorical inputs. These are expressed in the NeuralInputs element of the resulting PMML file.
- PMML does not support the censored variant of softmax.
- Given that
uses a single output node to represent binary classification, the resulting PMML file contains a discretizer with a threshold set to 0.5.**nnet**

*and export using the***nnet***package can be uploaded directly into ADAPA for real-time scoring or UPPI for big data scoring.***pmml**
## 0 Comments