Hi Vasu,
I understand that you want to use the weights of the model trained on an old dataset to retrain the same model architecture with a new dataset. You can follow the following steps to achieve the same:
- Create a new network with the same architecture as the old network.
- Initialize the weights of the new network with the trained weights from the old network using the "setwb" function.
new_network = setwb(new_network, weights); % Set the weights of the new network
3. Train the new network with the new data using the "trainNetwork" function.
new_network = trainNetwork(new_data, new_network); % Train the new network with new data
You can refer to the following MathWorks documentations to know more about "setwb" and "trainNetwork" functions respectively:
- https://www.mathworks.com/help/deeplearning/ref/setwb.html
- https://www.mathworks.com/help/deeplearning/ref/trainnetwork.html
I hope this helps.
