Porting this model for use in C#
Example of use in C#:
using AI.DataPrepaire.DataLoader.NNWBlockLoader; using AI.DataStructs.Algebraic; using AI.ONNX.NLP.Bert; string path = "SbertDistilAIFr"; // Загрузка модели Bert BertEmbedder embedder = BertEmbedder.FromPretrained(path); // Добавление последнего(линейного) слоя LinearLayerLoader linearLayer = LinearLayerLoader.LoadFromBinary(@$"{path}\1_Linear\model.aifr"); embedder.V2VBlocks.Add(linearLayer); // Векторизация предложений Vector vects1 = embedder.ForwardSBert("Kittens love milk"); Vector vects2 = embedder.ForwardSBert("Visual Studio Code (VS Code) — текстовый редактор, разработанный Microsoft для Windows, Linux и macOS."); Vector vects3 = embedder.ForwardSBert("Visual Studio Code, also commonly referred to as VS Code,[12] is a source-code editor developed by Microsoft for Windows, Linux and macOS."); Vector vects1_rus = embedder.ForwardSBert("Котята любят молоко"); // Рассчет близости между текстами Console.WriteLine(vects1.Cos(vects1_rus)); Console.WriteLine(vects1.Cos(vects2)); Console.WriteLine(vects2.Cos(vects3)); Console.WriteLine(vects1.Cos(vects3)); Console.ReadKey();
Network output
0.8312274467817271
0.18402859300814606
0.9556528309990797
0.14738954582438052
- Downloads last month
- 3
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.