Closed zeus527 closed 3 months ago
For simplicity, the embeddings in the dataset were obtained using all-MiniLM-L6-v2. The embedding of the model is 384 dimensions. This model we only carried out a simple processing, there are still some imperfections. A more reasonable approach will follow. If the embeddings in the data set do not meet the requirements, you can perform your own embeddings as needed. Welcome to communicate.
Issue was fixed. Thanks for the assistance
How can I to obtain the embeddings of paper and author in the TACM12K dataset. The code is not found in the rllm.