Some code cleanup

Author: alkampfergit

200_CSharpSemanticMemory/Helper/ExternalEmbeddingGenerator.cs

@@ -30,32 +30,37 @@ public ExternalEmbeddingGenerator(
             _embeddingGeneratorConfig = embeddingGeneratorConfig;
 
             _log = DefaultLogger<ExternalEmbeddingGenerator>.Instance;
+
+            ScanModel();
         }
 
-        private async Task GetDenseVectorSizeAsync()
+        private void ScanModel()
         {
             var client = _httpClientFactory.CreateClient();
             //do a post request to the model to get the dense vector size
-            var body = new { modelName = _embeddingGeneratorConfig.ModelName };
+            var body = new DimensionInput(_embeddingGeneratorConfig.ModelName);
             var request = new HttpRequestMessage(HttpMethod.Post, $"{_embeddingGeneratorConfig.Address.TrimEnd('/')}/dimensions")
             {
                 Content = new StringContent(JsonSerializer.Serialize(body), Encoding.UTF8, "application/json")
             };
-            var response = await client.GetAsync($"{_embeddingGeneratorConfig.Address.TrimEnd('/')}/dimensions");
+
+            var response = client.Send(request);
+
             //TODO: Proper error handling
             if (!response.IsSuccessStatusCode)
             {
-                throw new Exception("Failed to get dense vector size");
+                throw new Exception("Failed to get dimensions");
             }
-            var responseStream = await response.Content.ReadAsStreamAsync();
-            var embeddingInfo = await JsonSerializer.DeserializeAsync<EmbeddingInfo>(responseStream);
-            VectorSize = embeddingInfo.dimension;
+            var responseStream = response.Content.ReadAsStream();
+            var dimensionResult = JsonSerializer.Deserialize<DimensionResult>(responseStream);
+            this.VectorSize = dimensionResult.dimension;
+            this.MaxTokens = dimensionResult.maxSequenceLength;
         }
 
         /// <inheritdoc />
-        public int MaxTokens { get; private set; } = 4096;
+        public int MaxTokens { get; private set; } = 384;
 
-        public int VectorSize { get; private set; }
+        public int VectorSize { get; private set; } = 768;
 
         /// <inheritdoc />
         public int CountTokens(string text)
@@ -118,6 +123,8 @@ public async Task<Embedding> GenerateEmbeddingAsync(
         }
 
         public record SentenceInput(List<string> sentences, string modelName);
+        public record DimensionInput(string modelName);
+        public record DimensionResult(string model, int maxSequenceLength, int dimension);
         public record EmbeddingModel(List<double[]> embeddings, string model);
 
         private record EmbeddingInfo(string model, int dimension);

200_CSharpSemanticMemory/Samples/BookSample.cs

@@ -48,7 +48,10 @@ public async Task RunSample(string bookPdf)
             // Create sample pipeline with 4 files
             Console.WriteLine("* Defining pipeline with 4 files...");
             var pipeline = orchestrator
-                .PrepareNewDocumentUpload(index: "booksample", documentId: "booksample", new TagCollection { { "example", "books" } })
+                .PrepareNewDocumentUpload(
+                    index: "booksample", 
+                    documentId: "booksample", 
+                    new TagCollection { { "example", "books" } })
                 .AddUploadFile("file1", Path.GetFileName(bookPdf), bookPdf)
                 .Then("extract")
                 .Then("clean")

200_CSharpSemanticMemory/Samples/SBertSample.cs

@@ -46,15 +46,26 @@ public async Task RunSample(string bookPdf)
 
             if (useLocal)
             {
-                kernelMemoryBuilder.WithCustomTextGenerator(new LmStudioTextGeneration(httpClientFactory, new Uri("http://localhost:1234")));
+                kernelMemoryBuilder.WithCustomTextGenerator(
+                    new LmStudioTextGeneration(
+                        httpClientFactory,
+                        new Uri("http://localhost:1234")));
             }
             else
             {
                 kernelMemoryBuilder.WithAzureOpenAITextGeneration(chatConfig);
             }
 
+            var embeddingConfig = new ExternalEmbeddingGeneratorConfig()
+            {
+                ModelName = "msmarco-distilbert-base-v4"
+            };
+
             kernelMemoryBuilder
-                .WithCustomEmbeddingGenerator(new ExternalEmbeddingGenerator(httpClientFactory, new ExternalEmbeddingGeneratorConfig()))
+                .WithCustomEmbeddingGenerator(
+                    new ExternalEmbeddingGenerator(
+                        httpClientFactory,
+                        embeddingConfig))
                 .WithSimpleFileStorage(new SimpleFileStorageConfig()
                 {
                     Directory = "c:\\temp\\kmsbert\\storage",