Error initializing model :Microsoft.ML.OnnxRuntime.OnnxRuntimeException

Question

I encountered a problem while using ML.NET

An error occurred while calling ONNX for an open source project, and I don't know how to handle it

System.InvalidOperationException:“Error initializing model :Microsoft.ML.OnnxRuntime.OnnxRuntimeException: [ErrorCode:Fail] Deserialize tensor transformer.word_embeddings.weight failed.open file transformer.word_embeddings.weight fail, errcode = 2 - ϵͳ�Ҳ���ָ�����ļ���
   at Microsoft.ML.OnnxRuntime.NativeApiStatus.VerifySuccess(IntPtr nativeStatus)
   at Microsoft.ML.OnnxRuntime.InferenceSession.Init(String modelPath, SessionOptions options, PrePackedWeightsContainer prepackedWeightsContainer)
   at Microsoft.ML.OnnxRuntime.InferenceSession..ctor(String modelPath, SessionOptions options)
   at Microsoft.ML.Transforms.Onnx.OnnxModel..ctor(String modelFile, Nullable`1 gpuDeviceId, Boolean fallbackToCpu, Boolean ownModelFile, IDictionary`2 shapeDictionary, Int32 recursionLimit, Nullable`1 interOpNumThreads, Nullable`1 intraOpNumThreads)
   at Microsoft.ML.Transforms.Onnx.OnnxTransformer..ctor(IHostEnvironment env, Options options, Byte[] modelBytes)”

 public static EmbeddingInputEmbeds Gen_embedding(List<Int64> ids, bool useGpu)
    {
        if (ids.Count > 0)
        {
            EmbeddingInputids _input_id = new();
            MLContext _mlContext = new();
            _input_id.input_ids = ids.ToArray();
            var pipeline = _mlContext.Transforms.ApplyOnnxModel(modelFile: modelfile + "embedding.onnx",
                                                                gpuDeviceId: useGpu ? 0 : (int?)null,
                                                                inputColumnNames: new[] { "input_ids" },
                                                                outputColumnNames: new[] { "inputs_embeds" },
                                                                fallbackToCpu: true
            );

            IDataView data = _mlContext.Data.LoadFromEnumerable(new List<EmbeddingInputids>() { _input_id });
            var model = pipeline.Fit(data);
            IDataView transformedValues = model.Transform(data);
            IEnumerable<EmbeddingInputEmbeds> results = transformedValues.GetColumn<EmbeddingInputEmbeds>("inputs_embeds");
            if (results.Count() > 0)
                return results?.ElementAtOrDefault(0);
        }
        return new EmbeddingInputEmbeds();
    }

Onnx fromhttps://github.com/wangzhaode/ChatGLM-MNN/releases/tag/v0.4