'Extract weights and biases from Orange Data Mining
I am using Orange Datamining to train a model on the Iris data set. I used the stochastic gradient descent node to do the training and I am looking to extract the corresponding weights and biases of the network.
I am new to this so I could be wrong about SGD being a neural network. Is there a different node that is in fact a neural network? I need the exact weights and biases for a project I am working on and I don't care about how I get them. I know there is a python interface but I can't find the weights there either.
Is there a different software suite that would be more suited for this goal? I need software that makes training extremely easy and that outputs the weights and biases
Solution 1:[1]
An SGD widget is a GUI wrapper around Stochastic Gradient Descent from scikit-learn and it is not a neural network. No neural networks are currently supported in Orange. To get the model parameters for simpler models, like logistic regression, use the workflow where the data set is fed into the learner and coefficients are observed in the Data Table.
Solution 2:[2]
Neural Networks widget is currently supported by Orange , but there is no way to get the model parameters using Data Table
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|---|
| Solution 1 | blaz |
| Solution 2 | Hazem Abdelazim |