ImageGenerator | : heartpulse : | Runtime Evironment library

What you're referring here is what we call "Data Imbalance" problem in ML lingo. This refers to the fact when you have the number of observations in one class is much higher than the other class. You can think of a case where the ratio between 2 classes is something like 1:100. though there are no strict rules as to how imbalance the ratio should be.

In your case, the biggest imbalance between the 2 classes is 46, which doesn't seem that big of a difference in absolute term, being said that It also seems the sample size isn't quite big either in relative term.

The problem with data imbalance is you can think of it as a zero-sum game, where 2 classes keep pushing the decision boundary, so when you've 2 classes with quite the same amount of samples (It doesn't have to be strictly equal) then the decision boundary comes kind of at a "Nash equilibrium" state, means 2 classes are pushing the boundary at the same force, hence the boundary is at the middle and successfully discriminates the 2 classes, but when these forces become very un-equal then the majority class pushes the boundary as such that the minority class can't back it off, so the decision boundary fails to discriminate 2 classes...

(Note: The scenario mentioned above is highly superficial, imaginative (without much theory to back it off) and only described to build a perspective/intuition about the problem)

So, I would advise you to train the model with your current data the way it is, and see how it performs. From the theoretical point of view, It shouldn't be that bad, as your case isn't very bad. Nonetheless, ML is an empirical science, so you can also see what happens when some data imbalance technique is applied. there are techniques such as undersampling, oversampling. You can create data for the minority class (oversampling) or you can reduce the data samples of the majority class (undersampling). SMOTE is a popular oversampling technique and RUSBoost is a popular undersampling technique.

You need to use a DispatchGroup to wait for several async functions to complete. You need to make sure your calls to DispatchGroup.enter() and leave() are balanced (called the same number of times), since when you use dispatchGroup.notify(queue:), the closure will only be executed once leave() has been called for each enter().

You need to call enter for each element of your times array (which will contain the CMTime variables time1 to time7, then in the completion block of generateCGImagesAsynchronously(forTimes:, you call leave(). This ensures that the notify(queue:) will only execute its closure once generateCGImagesAsynchronously called its own completion for all images.

This error was solved by redefining a new image generator instead of simple_image_generator(). The simple_image_generator() worked well with the shape of the images (8 Bands) but did not cope well with the shape of the mask (1 band ).

During the execution, image_generator had 4 dimensions with [2,256,256,1] ( i.e. batch_size, (image size), bands) BUT mask_generator had 3 dimensions only vs. [2,256,256] (i.e. batch_size,(image size))

So reshaping the mask of [2,256,256] to [2,256,256, 1] solved the issue.

why did you use array in useState()?

You did every thing correct except one. When you are creating the data generator for testing (new_generator) you are not setting the shuffle=False. This way you can't reproduce the results with each run as the data generator will shuffle the data for each iteration.

I think the issue is that ImageDataGenerator expects an image with has a width, height, and the color channels (the most common being 3 channels for red, green, and blue). Since there's also a batch size the overall shape it expects is (batch size, width, height, channels). Your tensors are 260x260 but don't have the color channels. Are they grayscale images?

Per the documentation:

x: Sample data. Should have rank 4. In case of grayscale data, the channels axis should have value 1

So I think you just need to reshape your input adding an extra dimension at the end.

I figured out the issue. The issue is that you cannot correctly store a Tensor or a Numpy array in a Pandas Dataframe. I ended up having to manually create the image/tensor pairs by doing the following:

Your model and the generators not for multi class but binary classification. First you need to fix your model last layer to get output with class size. Second you need to fix the generators to use in multi class.

Found my mistake - explanation for future reference.

The Error origniates back in compile() where the first if-statement says: