Hello, @AndrewZhaoLuo @masahi Thanks for your answer.
@AndrewZhaoLuo Yes, I can definitely try to converting the model → onnx → relay. But I still wanna try on Pytorch for now. @masahi I have used "torch.jit.trace" to produce trace model, and it looks normal: > SqueezeBertForSequenceClassification( > original_name=SqueezeBertForSequenceClassification > (transformer): SqueezeBertModel( > original_name=SqueezeBertModel > (embeddings): SqueezeBertEmbeddings( > original_name=SqueezeBertEmbeddings > (word_embeddings): Embedding(original_name=Embedding) > (position_embeddings): Embedding(original_name=Embedding) > (token_type_embeddings): Embedding(original_name=Embedding) > (LayerNorm): LayerNorm(original_name=LayerNorm) > (dropout): Dropout(original_name=Dropout) > ).......... However, as I shown above, when I try to use > mod, params = relay.frontend.from_pytorch(traced_script_module, > input_infos=shape_dict) The following error: **operators are not implemented: ['prim::DictConstruct']** still shows up Note: I am using newest tvm version: tvm 0.9.dev32+gecd8a9ce3 --- [Visit Topic](https://discuss.tvm.apache.org/t/issue-converting-model-from-pytorch-to-relay-model/11538/4) to respond. You are receiving this because you enabled mailing list mode. To unsubscribe from these emails, [click here](https://discuss.tvm.apache.org/email/unsubscribe/9d9e98b2de4bc036d867a837a8d55091ce758cd8feaad24989ae2b5fd2c45faa).
