can you explain the function of the parameters in the runExperiment about incremental-learning

[khurramjaved96]

Yes, sure.
I'll update the code this week so it's more readable; there are some parameters corresponding to ideas that we tried that didn't work. I'll also remove those to make things more clear.

[khurramjaved96]

I've added slightly better description of each parameter; just run 'python run_experiment -h' to see what each parameter mean.

Let me know if something is not clear.

[fanxuansoft]

thank you

[fanxuansoft]

Hello,Your code is very useful to me, I have a question and hope to get your help.

I want to do incremental training of a deep convolutional neural network (CNN) model as new classes are added to the existing data. The CNN model is initially fully trained for classifying, say, 500 classes with 1 million images. Now, new data is available which has 50 new classes with 10000 images in addition to earlier 500 classes.

This addition of new classes in the data will be an ongoing occurrence every few days or every week.

Incremental training: I want to avoid training with full 1 million + 10000 images because of the training time. Any technique using the new 10000 images (50 classes) + tiny fraction of original 1 million (500 classes) will qualify as incremental training. I am aware of 'catastrophic forgetting' in NNs; however, I am looking for way to work around that.

I don't know if your code can solve my problem.and I hope you can give me some ideas.