NeuralNetClassifier
The Iris Flower Classifier, implemented with the NeuralNetClassifier algorithm, is based on the famous Iris database first introduced by Sir R.A. Fisher.
The Iris Flower Classifier, implemented with theNeuralNetClassifier algorithm, is based on the famous Iris database first introduced by Sir R.A. Fisher. The dataset is sourced directly from Fisher's paper, and it differs from the version in the UCI Machine Learning Repository, which contains two incorrect data points.
To use this model, select Model 4 in the Privasea client and copy your feature vector as the X vector. The client will locally encrypt this vector using your stored client key, and the encrypted data will be securely transmitted to the Privanetix nodes. Inference operations will occur in the encrypted domain on the Privanetix nodes, and the encrypted result will be sent back to your client. After decryption using your locally stored client key, you'll receive a three-dimensional vector representing the likelihood of the associated tags. For example, a result like "01, 0.3, 0.6" signifiesa classification as 2 (Virginica). You can then compare it with the tag in the provided output.
Feature Vector：
[
sepal length (cm)
sepal width (cm)
petal length (cm)
petal width (cm)
]
Output：
[
0, Setosa
1, Versicolor
2 Virginica
]