Spaces:

---

coreml-projects

/

transformers-to-coreml

like 97

Paused

Conversion Settings:

        Model: Helsinki-NLP/opus-mt-zh-en
        Task: feature-extraction
        Framework: pt
        Compute Units: ComputeUnit.ALL
        Precision: float32
        Tolerance: None
        Push to: None

        Error: You have to specify either decoder_input_ids or decoder_inputs_embeds
        

It seems MarianMTModel are encoder-decoder model, it should be converted separately
I found some similar issue here:
https://stackoverflow.com/questions/66109084/how-to-convert-huggingfaces-seq2seq-models-to-onnx-format/66117248#66117248