Closed mwawrzos closed 3 years ago
Issue reported by @qpjaada
When time and frequency masks are sampled, the position of the masks never fills the last column, row respectively: https://github.com/mlcommons/training/blob/master/rnn_speech_recognition/pytorch/common/data/features.py#L102 https://github.com/mlcommons/training/blob/master/rnn_speech_recognition/pytorch/common/data/features.py#L88
Issue reported by @qpjaada
When time and frequency masks are sampled, the position of the masks never fills the last column, row respectively: https://github.com/mlcommons/training/blob/master/rnn_speech_recognition/pytorch/common/data/features.py#L102 https://github.com/mlcommons/training/blob/master/rnn_speech_recognition/pytorch/common/data/features.py#L88