a simple implementation of self attention layer with the Frobenius norm penalty that produces flattened sentence embedding matrix for sentence representation learning tasks.
Based on the paper https://arxiv.org/abs/1703.03130
| Name | Name | Last commit date | ||
|---|---|---|---|---|
a simple implementation of self attention layer with the Frobenius norm penalty that produces flattened sentence embedding matrix for sentence representation learning tasks.
Based on the paper https://arxiv.org/abs/1703.03130