Open dongchirua opened 1 year ago
Hi there, we have released the embedding model of CodeT5+ 110M here, which achieves very competitive performance on multiple text-to-code retrieval tasks. The CodeT5+ 110M embedding model can extract 256-dimensional embeddings for both code and text.
您好,我们在这里发布了 CodeT5+ 110M 的嵌入模型,它在多个文本到代码检索任务上实现了极具竞争力的性能。CodeT5+ 110M 嵌入模型可以提取代码和文本的 256 维嵌入。
Hello, I want to implement an encoder - decoder program translation task through codeT5. But I need to manipulate the encoder generated embeddings. Is there any way to achieve this process?
Hi,
I wonder how I can use the encoder part to extract embedding.