Merge pull request #702 from martindevans/interruptible_async_model_load

Interruptible Async Model Loading With Progress Monitoring

LLama.Examples/Examples/BatchedExecutorFork.cs

@@ -19,7 +19,7 @@ public class BatchedExecutorFork
         string modelPath = UserSettings.GetModelPath();
 
         var parameters = new ModelParams(modelPath);
-        using var model = LLamaWeights.LoadFromFile(parameters);
+        using var model = await LLamaWeights.LoadFromFileAsync(parameters);
 
         var prompt = AnsiConsole.Ask("Prompt (or ENTER for default):", "Not many people know that");
 

LLama.Examples/Examples/BatchedExecutorGuidance.cs

@@ -19,7 +19,7 @@ public class BatchedExecutorGuidance
         string modelPath = UserSettings.GetModelPath();
 
         var parameters = new ModelParams(modelPath);
-        using var model = LLamaWeights.LoadFromFile(parameters);
+        using var model = await LLamaWeights.LoadFromFileAsync(parameters);
 
         var positivePrompt = AnsiConsole.Ask("Positive Prompt (or ENTER for default):", "My favourite colour is").Trim();
         var negativePrompt = AnsiConsole.Ask("Negative Prompt (or ENTER for default):", "I hate the colour red. My favourite colour is").Trim();

LLama.Examples/Examples/BatchedExecutorRewind.cs

@@ -20,7 +20,7 @@ public class BatchedExecutorRewind
         string modelPath = UserSettings.GetModelPath();
 
         var parameters = new ModelParams(modelPath);
-        using var model = LLamaWeights.LoadFromFile(parameters);
+        using var model = await LLamaWeights.LoadFromFileAsync(parameters);
 
         var prompt = AnsiConsole.Ask("Prompt (or ENTER for default):", "Not many people know that");
 

LLama.Examples/Examples/BatchedExecutorSaveAndLoad.cs

@@ -18,7 +18,7 @@ public class BatchedExecutorSaveAndLoad
         string modelPath = UserSettings.GetModelPath();
 
         var parameters = new ModelParams(modelPath);
-        using var model = LLamaWeights.LoadFromFile(parameters);
+        using var model = await LLamaWeights.LoadFromFileAsync(parameters);
 
         var prompt = AnsiConsole.Ask("Prompt (or ENTER for default):", "Not many people know that");
 

LLama.Examples/Examples/ChatChineseGB2312.cs

@@ -31,7 +31,7 @@ public class ChatChineseGB2312
             GpuLayerCount = 5,
             Encoding = Encoding.UTF8
         };
-        using var model = LLamaWeights.LoadFromFile(parameters);
+        using var model = await LLamaWeights.LoadFromFileAsync(parameters);
         using var context = model.CreateContext(parameters);
         var executor = new InteractiveExecutor(context);
 

LLama.Examples/Examples/ChatSessionStripRoleName.cs

@@ -15,11 +15,11 @@ public class ChatSessionStripRoleName
             Seed = 1337,
             GpuLayerCount = 5
         };
-        using var model = LLamaWeights.LoadFromFile(parameters);
+        using var model = await LLamaWeights.LoadFromFileAsync(parameters);
         using var context = model.CreateContext(parameters);
         var executor = new InteractiveExecutor(context);
 
-        var chatHistoryJson = File.ReadAllText("Assets/chat-with-bob.json");
+        var chatHistoryJson = await File.ReadAllTextAsync("Assets/chat-with-bob.json");
         ChatHistory chatHistory = ChatHistory.FromJson(chatHistoryJson) ?? new ChatHistory();
 
         ChatSession session = new(executor, chatHistory);

LLama.Examples/Examples/ChatSessionWithHistory.cs

@@ -13,7 +13,7 @@ public class ChatSessionWithHistory
             Seed = 1337,
             GpuLayerCount = 5
         };
-        using var model = LLamaWeights.LoadFromFile(parameters);
+        using var model = await LLamaWeights.LoadFromFileAsync(parameters);
         using var context = model.CreateContext(parameters);
         var executor = new InteractiveExecutor(context);
 

LLama.Examples/Examples/ChatSessionWithRestart.cs

@@ -13,11 +13,11 @@ public class ChatSessionWithRestart
             Seed = 1337,
             GpuLayerCount = 5
         };
-        using var model = LLamaWeights.LoadFromFile(parameters);
+        using var model = await LLamaWeights.LoadFromFileAsync(parameters);
         using var context = model.CreateContext(parameters);
         var executor = new InteractiveExecutor(context);
 
-        var chatHistoryJson = File.ReadAllText("Assets/chat-with-bob.json");
+        var chatHistoryJson = await File.ReadAllTextAsync("Assets/chat-with-bob.json");
         ChatHistory chatHistory = ChatHistory.FromJson(chatHistoryJson) ?? new ChatHistory();
         ChatSession prototypeSession = 
             await ChatSession.InitializeSessionFromHistoryAsync(executor, chatHistory);

LLama.Examples/Examples/ChatSessionWithRoleName.cs

@@ -13,11 +13,11 @@ public class ChatSessionWithRoleName
             Seed = 1337,
             GpuLayerCount = 5
         };
-        using var model = LLamaWeights.LoadFromFile(parameters);
+        using var model = await LLamaWeights.LoadFromFileAsync(parameters);
         using var context = model.CreateContext(parameters);
         var executor = new InteractiveExecutor(context);
 
-        var chatHistoryJson = File.ReadAllText("Assets/chat-with-bob.json");
+        var chatHistoryJson = await File.ReadAllTextAsync("Assets/chat-with-bob.json");
         ChatHistory chatHistory = ChatHistory.FromJson(chatHistoryJson) ?? new ChatHistory();
 
         ChatSession session = new(executor, chatHistory);

LLama.Examples/Examples/CodingAssistant.cs

@@ -29,7 +29,7 @@
             {
                 ContextSize = 4096
             };
-            using var model = LLamaWeights.LoadFromFile(parameters);
+            using var model = await LLamaWeights.LoadFromFileAsync(parameters);
             using var context = model.CreateContext(parameters);
             var executor = new InstructExecutor(context, InstructionPrefix, InstructionSuffix, null);
 

LLama.Examples/Examples/GrammarJsonResponse.cs

@@ -9,7 +9,7 @@ namespace LLama.Examples.Examples
         {
             string modelPath = UserSettings.GetModelPath();
 
-            var gbnf = File.ReadAllText("Assets/json.gbnf").Trim();
+            var gbnf = (await File.ReadAllTextAsync("Assets/json.gbnf")).Trim();
             var grammar = Grammar.Parse(gbnf, "root");
 
             var parameters = new ModelParams(modelPath)
@@ -17,7 +17,7 @@ namespace LLama.Examples.Examples
                 Seed = 1337,
                 GpuLayerCount = 5
             };
-            using var model = LLamaWeights.LoadFromFile(parameters);
+            using var model = await LLamaWeights.LoadFromFileAsync(parameters);
             var ex = new StatelessExecutor(model, parameters);
 
             Console.ForegroundColor = ConsoleColor.Yellow;

LLama.Examples/Examples/InstructModeExecute.cs

@@ -9,14 +9,14 @@ namespace LLama.Examples.Examples
         {
             string modelPath = UserSettings.GetModelPath();
 
-            var prompt = File.ReadAllText("Assets/dan.txt").Trim();
+            var prompt = (await File.ReadAllTextAsync("Assets/dan.txt")).Trim();
 
             var parameters = new ModelParams(modelPath)
             {
                 Seed = 1337,
                 GpuLayerCount = 5
             };
-            using var model = LLamaWeights.LoadFromFile(parameters);
+            using var model = await LLamaWeights.LoadFromFileAsync(parameters);
             using var context = model.CreateContext(parameters);
             var executor = new InstructExecutor(context);
 

LLama.Examples/Examples/InteractiveModeExecute.cs

@@ -16,7 +16,7 @@ namespace LLama.Examples.Examples
                 Seed = 1337,
                 GpuLayerCount = 5
             };
-            using var model = LLamaWeights.LoadFromFile(parameters);
+            using var model = await LLamaWeights.LoadFromFileAsync(parameters);
             using var context = model.CreateContext(parameters);
             var ex = new InteractiveExecutor(context);
 

LLama.Examples/Examples/LlavaInteractiveModeExecute.cs

@@ -20,7 +20,7 @@ namespace LLama.Examples.Examples
 
             var parameters = new ModelParams(modelPath);
 
-            using var model = LLamaWeights.LoadFromFile(parameters);
+            using var model = await LLamaWeights.LoadFromFileAsync(parameters);
             using var context = model.CreateContext(parameters);
             
             // Llava Init

LLama.Examples/Examples/LoadAndSaveSession.cs

@@ -15,7 +15,7 @@ namespace LLama.Examples.Examples
                 Seed = 1337,
                 GpuLayerCount = 5
             };
-            using var model = LLamaWeights.LoadFromFile(parameters);
+            using var model = await LLamaWeights.LoadFromFileAsync(parameters);
             using var context = model.CreateContext(parameters);
             var ex = new InteractiveExecutor(context);
 

LLama.Examples/Examples/LoadAndSaveState.cs

@@ -16,7 +16,7 @@ namespace LLama.Examples.Examples
                 Seed = 1337,
                 GpuLayerCount = 5
             };
-            using var model = LLamaWeights.LoadFromFile(parameters);
+            using var model = await LLamaWeights.LoadFromFileAsync(parameters);
             using var context = model.CreateContext(parameters);
             var ex = new InteractiveExecutor(context);
 

LLama.Examples/Examples/SemanticKernelChat.cs

@@ -16,7 +16,7 @@ namespace LLama.Examples.Examples
 
             // Load weights into memory
             var parameters = new ModelParams(modelPath);
-            using var model = LLamaWeights.LoadFromFile(parameters);
+            using var model = await LLamaWeights.LoadFromFileAsync(parameters);
             var ex = new StatelessExecutor(model, parameters);
 
             var chatGPT = new LLamaSharpChatCompletion(ex);

LLama.Examples/Examples/SemanticKernelMemory.cs

@@ -23,7 +23,7 @@ namespace LLama.Examples.Examples
                 Embeddings = true
             };
 
-            using var model = LLamaWeights.LoadFromFile(parameters);
+            using var model = await LLamaWeights.LoadFromFileAsync(parameters);
             var embedding = new LLamaEmbedder(model, parameters);
 
             Console.WriteLine("====================================================");

LLama.Examples/Examples/SemanticKernelPrompt.cs

@@ -19,7 +19,7 @@ namespace LLama.Examples.Examples
 
             // Load weights into memory
             var parameters = new ModelParams(modelPath);
-            using var model = LLamaWeights.LoadFromFile(parameters);
+            using var model = await LLamaWeights.LoadFromFileAsync(parameters);
             var ex = new StatelessExecutor(model, parameters);
 
             var builder = Kernel.CreateBuilder();

LLama.Examples/Examples/StatelessModeExecute.cs

@@ -14,7 +14,7 @@ namespace LLama.Examples.Examples
                 Seed = 1337,
                 GpuLayerCount = 5
             };
-            using var model = LLamaWeights.LoadFromFile(parameters);
+            using var model = await LLamaWeights.LoadFromFileAsync(parameters);
             var ex = new StatelessExecutor(model, parameters);
 
             Console.ForegroundColor = ConsoleColor.Yellow;

LLama.Examples/Examples/TalkToYourself.cs

@@ -12,7 +12,7 @@ namespace LLama.Examples.Examples
 
             // Load weights into memory
             var @params = new ModelParams(modelPath);
-            using var weights = LLamaWeights.LoadFromFile(@params);
+            using var weights = await LLamaWeights.LoadFromFileAsync(@params);
 
             // Create 2 contexts sharing the same weights
             using var aliceCtx = weights.CreateContext(@params);

LLama/LLamaWeights.cs

@@ -1,7 +1,10 @@
 using System;
 using System.Collections.Generic;
 using System.Text;
+using System.Threading;
+using System.Threading.Tasks;
 using LLama.Abstractions;
+using LLama.Exceptions;
 using LLama.Extensions;
 using LLama.Native;
 using Microsoft.Extensions.Logging;
@@ -84,6 +87,104 @@ namespace LLama
             return new LLamaWeights(weights);
         }
 
+        /// <summary>
+        /// Load weights into memory
+        /// </summary>
+        /// <param name="params">Parameters to use to load the model</param>
+        /// <param name="token">A cancellation token that can interrupt model loading</param>
+        /// <param name="progressReporter">Receives progress updates as the model loads (0 to 1)</param>
+        /// <returns></returns>
+        /// <exception cref="LoadWeightsFailedException">Thrown if weights failed to load for any reason. e.g. Invalid file format or loading cancelled.</exception>
+        /// <exception cref="OperationCanceledException">Thrown if the cancellation token is cancelled.</exception>
+        public static async Task<LLamaWeights> LoadFromFileAsync(IModelParams @params, CancellationToken token = default, IProgress<float>? progressReporter = null)
+        {
+            // don't touch the @params object inside the task, it might be changed
+            // externally! Save a copy of everything that we need later.
+            var modelPath = @params.ModelPath;
+            var loraBase = @params.LoraBase;
+            var loraAdapters = @params.LoraAdapters.ToArray();
+
+            // Determine the range to report for model loading. llama.cpp reports 0-1, but we'll remap that into a
+            // slightly smaller range to allow some space for reporting LoRA loading too.
+            var modelLoadProgressRange = 1f;
+            if (loraAdapters.Length > 0)
+                modelLoadProgressRange = 0.9f;
+
+            using (@params.ToLlamaModelParams(out var lparams))
+            {
+#if !NETSTANDARD2_0
+                // Overwrite the progress callback with one which polls the cancellation token and updates the progress object
+                if (token.CanBeCanceled || progressReporter != null)
+                {
+                    var internalCallback = lparams.progress_callback;
+                    lparams.progress_callback = (progress, ctx) =>
+                    {
+                        // Update the progress reporter (remapping the value into the smaller range).
+                        progressReporter?.Report(Math.Clamp(progress, 0, 1) * modelLoadProgressRange);
+
+                        // If the user set a callback in the model params, call that and see if we should cancel
+                        if (internalCallback != null && !internalCallback(progress, ctx))
+                            return false;
+
+                        // Check the cancellation token
+                        if (token.IsCancellationRequested)
+                            return false;
+
+                        return true;
+                    };
+                }
+#endif
+
+                var model = await Task.Run(() =>
+                {
+                    try
+                    {
+                        // Load the model
+                        var weights = SafeLlamaModelHandle.LoadFromFile(modelPath, lparams);
+
+                        // Apply the LoRA adapters
+                        for (var i = 0; i < loraAdapters.Length; i++)
+                        {
+                            // Interrupt applying LoRAs if the token is cancelled
+                            if (token.IsCancellationRequested)
+                            {
+                                weights.Dispose();
+                                token.ThrowIfCancellationRequested();
+                            }
+
+                            // Don't apply invalid adapters
+                            var adapter = loraAdapters[i];
+                            if (string.IsNullOrEmpty(adapter.Path))
+                                continue;
+                            if (adapter.Scale <= 0)
+                                continue;
+
+                            weights.ApplyLoraFromFile(adapter.Path, adapter.Scale, loraBase);
+
+                            // Report progress. Model loading reported progress from 0 -> 0.9, use
+                            // the last 0.1 to represent all of the LoRA adapters being applied.
+                            progressReporter?.Report(0.9f + (0.1f / loraAdapters.Length) * (i + 1));
+                        }
+
+                        // Update progress reporter to indicate completion
+                        progressReporter?.Report(1);
+
+                        return new LLamaWeights(weights);
+                    }
+                    catch (LoadWeightsFailedException)
+                    {
+                        // Convert a LoadWeightsFailedException into a cancellation exception if possible.
+                        token.ThrowIfCancellationRequested();
+
+                        // Ok the weights failed to load for some reason other than cancellation.
+                        throw;
+                    }
+                }, token);
+
+                return model;
+            }
+        }
+
         /// <inheritdoc />
         public void Dispose()
         {

LLama/Native/LLamaContextParams.cs

@@ -8,6 +8,8 @@ namespace LLama.Native
     /// </summary>
     /// <param name="progress"></param>
     /// <param name="ctx"></param>
+    /// <returns>If the provided progress_callback returns true, model loading continues.
+    /// If it returns false, model loading is immediately aborted.</returns>
     /// <remarks>llama_progress_callback</remarks>
     public delegate bool LlamaProgressCallback(float progress, IntPtr ctx);
 

LLama/Native/LLamaModelParams.cs

@@ -38,7 +38,7 @@ namespace LLama.Native
         // as NET Framework 4.8 does not play nice with the LlamaProgressCallback type
         public IntPtr progress_callback;
 #else
-        public LlamaProgressCallback progress_callback;
+        public LlamaProgressCallback? progress_callback;
 #endif
 
         /// <summary>

LLama/Native/SafeLlamaModelHandle.cs

@@ -120,8 +120,11 @@ namespace LLama.Native
                 if (!fs.CanRead)
                     throw new InvalidOperationException($"Model file '{modelPath}' is not readable");
 
-            return llama_load_model_from_file(modelPath, lparams)
-                ?? throw new LoadWeightsFailedException(modelPath);
+            var handle = llama_load_model_from_file(modelPath, lparams);
+            if (handle.IsInvalid)
+                throw new LoadWeightsFailedException(modelPath);
+
+            return handle;
         }
 
         #region native API