I have read on Internet that epochs is used to give the time for the model to converge but I don't know how ? . I was thinking that epochs is used because to train the model sufficient times . How does model convergence relates with epochs . Also tell me that why epochs is useful ?
Topic epochs machine-learning-model deep-learning
Category Data Science
I don't know if I understood correctly the question, but basically, epochs are the unit of measure of the time spent on training the machine learning models.
An epoch is conventionally considered as a full pass on the dataset, meaning that for each epoch, you feed the data to the model from the beginning to the end of the dataset. Once an epoch is complete, you can start again another epoch training your model from the beginning of the dataset to the end, and so on and so forth.
The convergence of the performance of the models is of course dependant on the epochs, because usually what happens is that the more epochs you train your model, the less noisy (more stable) will be the predictions.
Hope I answered your question.
Geeks Mental is a community that publishes articles and tutorials about Web, Android, Data Science, new techniques and Linux security.