This project will involve benchmarking state of the art methods for time series classification on the new MONSTER benchmark datasets [1, 2, 3].  Currently almost all benchmarking in time series classification is performed on the (almost all very small) datasets in the UCR and UEA archives.  This is particularly unsuitable for deep learning models which are low bias models and ideally trained using large quantities of data.  The "true" performance of current deep learning methods for time series classification is unknown outside of the UCR/UEA datasets.  Most deep learning models for times series classification are configured to be trained on tiny quantities of data.  A lot of published work on deep learning for time series classification has serious methodological flaws (e.g., directly or indirectly optimising on the test data).  This project would involve establishing a sound training setup for deep learning models on large quantities of training data, and training and deep learning models for time series classification on the much larger datasets in the new MONSTER benchmark, in order to establish their "true" performance on large datasets.