Hello, thanks for you wonderful work! I would like to ask how to generate the word embeddings in your paper, can you provide some instructions or code for that?
As you can see in the paper and in the file embeddings/coco.txt.
A single category is fed directly into the word vector model to get a 300-dimensional result.
If a category contains multiple words, they will be sent to the word vector model separately, and the average value will be used as the result, such as traffic and light; stop and sign.
All output results in each dataset will eventually be normalized.
The complete instructions or code is being compiled soon.
Hello, thanks for you wonderful work! I would like to ask how to generate the word embeddings in your paper, can you provide some instructions or code for that?