Epilepsy affects more than 50 million people worldwide, making it one of the world’s most prevalent neurological diseases. The main symptom of epilepsy is seizures, which occur abruptly and can cause serious injury or death. The ability to predict the occurrence of an epileptic seizure could alleviate many risks and stresses people with epilepsy face. We have taken several approaches to predict the occurrence of a seizure using deep learning. Firstly, we developed a supervised model to identify pre-seizure EEG from normal EEG. Then, we developed an unsupervised approach where the model is trained on just the normal EEG, and pre-seizure EEG is identified as an anomalous event indicating the onset of a seizure. These models were trained and evaluated on two public EEG databases. We have found that both supervised and unsupervised approaches are feasible;​ however, their performance varies depending on the patient and the approach and architecture.