feat: rendering chat_template

namchuai · namchuai · commit 65f9790aacb4 · 2024-12-20T01:39:36.000+07:00
diff --git a/engine/cli/commands/chat_completion_cmd.cc b/engine/cli/commands/chat_completion_cmd.cc
@@ -151,9 +151,8 @@ void ChatCompletionCmd::Exec(const std::string& host, int port,
       json_data["model"] = model_handle;
       json_data["stream"] = true;
 
-      std::string json_payload = json_data.toStyledString();
-
-      curl_easy_setopt(curl, CURLOPT_POSTFIELDS, json_payload.c_str());
+      curl_easy_setopt(curl, CURLOPT_POSTFIELDS,
+                       json_data.toStyledString().c_str());
 
       std::string ai_chat;
       StreamingCallback callback;
diff --git a/engine/common/model_metadata.h b/engine/common/model_metadata.h
@@ -0,0 +1,29 @@
+#pragma once
+
+#include "common/tokenizer.h"
+#include <sstream>
+
+struct ModelMetadata {
+  uint32_t version;
+  uint64_t tensor_count;
+  uint64_t metadata_kv_count;
+  std::unique_ptr<Tokenizer> tokenizer;
+
+  std::string ToString() const {
+    std::ostringstream ss;
+    ss << "ModelMetadata {\n"
+       << "version: " << version << "\n"
+       << "tensor_count: " << tensor_count << "\n"
+       << "metadata_kv_count: " << metadata_kv_count << "\n"
+       << "tokenizer: ";
+    
+    if (tokenizer) {
+        ss << "\n" << tokenizer->ToString();
+    } else {
+        ss << "null";
+    }
+    
+    ss << "\n}";
+    return ss.str();
+  }
+};
diff --git a/engine/common/tokenizer.h b/engine/common/tokenizer.h
@@ -0,0 +1,68 @@
+#pragma once
+
+#include <sstream>
+#include <string>
+
+struct Tokenizer {
+  std::string eos_token = "";
+  bool add_eos_token = true;
+
+  std::string bos_token = "";
+  bool add_bos_token = true;
+
+  std::string unknown_token = "";
+  std::string padding_token = "";
+
+  std::string chat_template = "";
+
+  // Helper function for common fields
+  std::string BaseToString() const {
+    std::ostringstream ss;
+    ss << "eos_token: \"" << eos_token << "\"\n"
+       << "add_eos_token: " << (add_eos_token ? "true" : "false") << "\n"
+       << "bos_token: \"" << bos_token << "\"\n"
+       << "add_bos_token: " << (add_bos_token ? "true" : "false") << "\n"
+       << "unknown_token: \"" << unknown_token << "\"\n"
+       << "padding_token: \"" << padding_token << "\"\n"
+       << "chat_template: \"" << chat_template << "\"";
+    return ss.str();
+  }
+
+  virtual ~Tokenizer() = default;
+
+  virtual std::string ToString() = 0;
+};
+
+struct GgufTokenizer : public Tokenizer {
+  std::string pre = "";
+
+  ~GgufTokenizer() override = default;
+
+  std::string ToString() override {
+    std::ostringstream ss;
+    ss << "GgufTokenizer {\n";
+    // Add base class members
+    ss << BaseToString() << "\n";
+    // Add derived class members
+    ss << "pre: \"" << pre << "\"\n";
+    ss << "}";
+    return ss.str();
+  }
+};
+
+struct SafeTensorTokenizer : public Tokenizer {
+  bool add_prefix_space = true;
+
+  ~SafeTensorTokenizer() = default;
+
+  std::string ToString() override {
+    std::ostringstream ss;
+    ss << "SafeTensorTokenizer {\n";
+    // Add base class members
+    ss << BaseToString() << "\n";
+    // Add derived class members
+    ss << "add_prefix_space: " << (add_prefix_space ? "true" : "false") << "\n";
+    ss << "}";
+    return ss.str();
+  }
+};
diff --git a/engine/controllers/engines.cc b/engine/controllers/engines.cc
@@ -3,6 +3,7 @@
 #include "utils/archive_utils.h"
 #include "utils/cortex_utils.h"
 #include "utils/engine_constants.h"
+#include "utils/jinja_utils.h"
 #include "utils/logging_utils.h"
 #include "utils/string_utils.h"
 
@@ -20,6 +21,41 @@ std::string NormalizeEngine(const std::string& engine) {
 };
 }  // namespace
 
+void Engines::TestJinja(
+    const HttpRequestPtr& req,
+    std::function<void(const HttpResponsePtr&)>&& callback) {
+  auto body = req->getJsonObject();
+  if (body == nullptr) {
+    Json::Value ret;
+    ret["message"] = "Body can't be empty";
+    auto resp = cortex_utils::CreateCortexHttpJsonResponse(ret);
+    resp->setStatusCode(k400BadRequest);
+    callback(resp);
+    return;
+  }
+
+  auto jinja = body->get("jinja", "").asString();
+  auto data = body->get("data", {});
+  auto bos_token = data.get("bos_token", "").asString();
+  auto eos_token = data.get("eos_token", "").asString();
+
+  auto rendered_data = jinja::RenderTemplate(jinja, data, bos_token, eos_token);
+
+  if (rendered_data.has_error()) {
+    Json::Value ret;
+    ret["message"] = rendered_data.error();
+    auto resp = cortex_utils::CreateCortexHttpJsonResponse(ret);
+    resp->setStatusCode(k400BadRequest);
+    callback(resp);
+    return;
+  }
+  // TODO: namh recheck all the api using this. because we have an issue with Germany locale before.
+  auto resp = HttpResponse::newHttpResponse();
+  resp->setBody(rendered_data.value());
+  resp->setContentTypeCode(drogon::CT_TEXT_PLAIN);
+  callback(resp);
+}
+
 void Engines::ListEngine(
     const HttpRequestPtr& req,
     std::function<void(const HttpResponsePtr&)>&& callback) const {
diff --git a/engine/controllers/engines.h b/engine/controllers/engines.h
@@ -12,6 +12,8 @@ class Engines : public drogon::HttpController<Engines, false> {
  public:
   METHOD_LIST_BEGIN
 
+  ADD_METHOD_TO(Engines::TestJinja, "/v1/jinja", Options, Post);
+
   // install engine
   METHOD_ADD(Engines::InstallEngine, "/{1}/install", Options, Post);
   ADD_METHOD_TO(Engines::InstallEngine, "/v1/engines/{1}/install", Options,
@@ -110,6 +112,9 @@ class Engines : public drogon::HttpController<Engines, false> {
       std::function<void(const HttpResponsePtr&)>&& callback,
       const std::string& engine) const;
 
+  void TestJinja(const HttpRequestPtr& req,
+                 std::function<void(const HttpResponsePtr&)>&& callback);
+
   void LoadEngine(const HttpRequestPtr& req,
                   std::function<void(const HttpResponsePtr&)>&& callback,
                   const std::string& engine);
diff --git a/engine/controllers/server.cc b/engine/controllers/server.cc
@@ -3,7 +3,6 @@
 #include "trantor/utils/Logger.h"
 #include "utils/cortex_utils.h"
 #include "utils/function_calling/common.h"
-#include "utils/http_util.h"
 
 using namespace inferences;
 
@@ -27,6 +26,14 @@ void server::ChatCompletion(
     std::function<void(const HttpResponsePtr&)>&& callback) {
   LOG_DEBUG << "Start chat completion";
   auto json_body = req->getJsonObject();
+  if (json_body == nullptr) {
+    Json::Value ret;
+    ret["message"] = "Body can't be empty";
+    auto resp = cortex_utils::CreateCortexHttpJsonResponse(ret);
+    resp->setStatusCode(k400BadRequest);
+    callback(resp);
+    return;
+  }
   bool is_stream = (*json_body).get("stream", false).asBool();
   auto model_id = (*json_body).get("model", "invalid_model").asString();
   auto engine_type = [this, &json_body]() -> std::string {
diff --git a/engine/services/engine_service.h b/engine/services/engine_service.h
@@ -4,7 +4,6 @@
 #include <mutex>
 #include <optional>
 #include <string>
-#include <string_view>
 #include <unordered_map>
 #include <vector>
 
@@ -17,7 +16,6 @@
 #include "utils/cpuid/cpu_info.h"
 #include "utils/dylib.h"
 #include "utils/dylib_path_manager.h"
-#include "utils/engine_constants.h"
 #include "utils/github_release_utils.h"
 #include "utils/result.hpp"
 #include "utils/system_info_utils.h"
@@ -48,10 +46,6 @@ class EngineService : public EngineServiceI {
   struct EngineInfo {
     std::unique_ptr<cortex_cpp::dylib> dl;
     EngineV engine;
-#if defined(_WIN32)
-    DLL_DIRECTORY_COOKIE cookie;
-    DLL_DIRECTORY_COOKIE cuda_cookie;
-#endif
   };
 
   std::mutex engines_mutex_;
diff --git a/engine/services/inference_service.cc b/engine/services/inference_service.cc
@@ -1,7 +1,10 @@
 #include "inference_service.h"
 #include <drogon/HttpTypes.h>
 #include "utils/engine_constants.h"
+#include "utils/file_manager_utils.h"
 #include "utils/function_calling/common.h"
+#include "utils/gguf_metadata_reader.h"
+#include "utils/jinja_utils.h"
 
 namespace services {
 cpp::result<void, InferResult> InferenceService::HandleChatCompletion(
@@ -24,6 +27,41 @@ cpp::result<void, InferResult> InferenceService::HandleChatCompletion(
     return cpp::fail(std::make_pair(stt, res));
   }
 
+  {
+    if (json_body->isMember("files") && !(*json_body)["files"].empty()) {
+      auto file = (*json_body)["files"][0].asString();
+      auto model_metadata_res = cortex_utils::ReadGgufMetadata(
+          file_manager_utils::ToAbsoluteCortexDataPath(
+              std::filesystem::path(file)));
+      if (model_metadata_res.has_value()) {
+        auto metadata = model_metadata_res.value().get();
+        if (!metadata->tokenizer->chat_template.empty()) {
+          auto messages = (*json_body)["messages"];
+          Json::Value messages_jsoncpp(Json::arrayValue);
+          for (auto message : messages) {
+            messages_jsoncpp.append(message);
+          }
+
+          Json::Value tools(Json::arrayValue);
+          Json::Value template_data_json;
+          template_data_json["messages"] = messages_jsoncpp;
+          // template_data_json["tools"] = tools;
+
+          auto prompt_result = jinja::RenderTemplate(
+              metadata->tokenizer->chat_template, template_data_json,
+              metadata->tokenizer->bos_token, metadata->tokenizer->eos_token);
+          if (prompt_result.has_value()) {
+            (*json_body)["prompt"] = prompt_result.value();
+          } else {
+            CTL_ERR("Failed to render prompt: " + prompt_result.error());
+          }
+        }
+      }
+    }
+  }
+
+  CTL_INF("Prompt is: " + json_body->get("prompt", "").asString());
+
   auto cb = [q, tool_choice](Json::Value status, Json::Value res) {
     if (!tool_choice.isNull()) {
       res["tool_choice"] = tool_choice;
@@ -297,4 +335,4 @@ bool InferenceService::HasFieldInReq(std::shared_ptr<Json::Value> json_body,
   }
   return true;
 }
-}  // namespace services
+}  // namespace services
diff --git a/engine/services/inference_service.h b/engine/services/inference_service.h
@@ -5,8 +5,9 @@
 #include <queue>
 #include "services/engine_service.h"
 #include "utils/result.hpp"
-#include "extensions/remote-engine/remote_engine.h"
+
 namespace services {
+
 // Status and result
 using InferResult = std::pair<Json::Value, Json::Value>;
 
diff --git a/engine/services/model_service.cc b/engine/services/model_service.cc
@@ -10,7 +10,6 @@
 #include "database/models.h"
 #include "hardware_service.h"
 #include "utils/cli_selection_utils.h"
-#include "utils/cortex_utils.h"
 #include "utils/engine_constants.h"
 #include "utils/file_manager_utils.h"
 #include "utils/huggingface_utils.h"
diff --git a/engine/test/components/test_gguf_parser.cc b/engine/test/components/test_gguf_parser.cc
diff --git a/engine/utils/chat-template.hpp b/engine/utils/chat-template.hpp
diff --git a/engine/utils/gguf_metadata_reader.h b/engine/utils/gguf_metadata_reader.h
diff --git a/engine/utils/jinja_utils.h b/engine/utils/jinja_utils.h
diff --git a/engine/utils/minja.hpp b/engine/utils/minja.hpp
