Skip to content

Latest commit

 

History

History
11 lines (8 loc) · 470 Bytes

attention_weights.md

File metadata and controls

11 lines (8 loc) · 470 Bytes

Attention Weight Analysis

Jupyter Notebook

A Jupyter Notebook was created to visualize the attention weights of the TransCoder-ST model. It is available here.

Make sure you used the created conda environment c2c-translation. To add it as a kernel for your notebook, run the following commands:

conda install -c anaconda ipykernel
python -m ipykernel install --user --name=c2c-translation