Why is the kernel dying on Jupyter?

Question

Python

I train the model on doc2vec during training, an error occurs that the core has died.

The kernel appears to have died. It will restart automatically.

I understand that such an error appears when there is not enough memory, how to write the code correctly?

corpus = ''
normal = Normalizer()
if __name__ == "__main__":
    for i, file in enumerate(files):
        try:
            with open('.....', 'r') as text:
                corpus += text.read()

            if i % 100 == 0 or i == len(files):
                sentences = normal.normalize(corpus)
                documents = [TaggedDocument(sentences, [i])]
                
                model.train(documents, total_examples=model.corpus_count, epochs=model.epochs)
                model.build_vocab(documents, update=True)
                model.save('model.model')
                corpus = ''

            if i % 1000 == 0:
                    print('Loading {0}/{1}, '.format(i, len(files)))
        except BaseException:
            print('ERROR {0}'.format(file))

The error appears on the line:

model.train(documents, total_examples=model.corpus_count, epochs=model.epochs)

since I have a lot of data, I divide them into groups of 100 files and train the model on them, after which I take the next group.
yes, yes, I'm stupid, you can do it without sarcasm and wit, I really need help.
I am just learning.
thanks in advance.