Fix toctree and multiagent rag system

huggingface · Jan 6, 2025 · b75779e · b75779e
1 parent 7a81cb9
commit b75779e
Show file tree

Hide file tree

Showing 3 changed files with 51 additions and 34 deletions.
diff --git a/notebooks/en/_toctree.yml b/notebooks/en/_toctree.yml
@@ -120,8 +120,6 @@
           title: Build an agent with tool-calling superpowers using Transformers Agents
         - local: agent_rag
           title: Agentic RAG - turbocharge your RAG with query reformulation and self-query
-        - local: agent_change_llm
-          title: Create a Transformers Agent from any LLM inference provider
         - local: agent_text_to_sql
           title: Agent for Text-to-SQL with automatic error correction
         - local: agent_data_analyst

diff --git a/notebooks/en/multiagent_rag_system.ipynb b/notebooks/en/multiagent_rag_system.ipynb
@@ -47,30 +47,52 @@
     },
     {
       "cell_type": "code",
-      "execution_count": null,
+      "execution_count": 10,
       "metadata": {
         "colab": {
           "base_uri": "https://localhost:8080/"
         },
         "id": "8rVK5wUFrNUI",
         "outputId": "acfadab7-6fbf-4b8e-e6ce-7597a1b9687d"
       },
-      "outputs": [],
+      "outputs": [
+        {
+          "name": "stderr",
+          "output_type": "stream",
+          "text": [
+            "huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\n",
+            "To disable this warning, you can either:\n",
+            "\t- Avoid using `tokenizers` before the fork if possible\n",
+            "\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\n"
+          ]
+        }
+      ],
       "source": [
-        "!pip install -q git+https://github.com/huggingface/transformers.git#egg=transformers[agents]"
+        "!pip install -q smolagents"
       ]
     },
     {
       "cell_type": "code",
-      "execution_count": null,
+      "execution_count": 11,
       "metadata": {
         "colab": {
           "base_uri": "https://localhost:8080/"
         },
         "id": "62DKn6XJB0MW",
         "outputId": "c03b6378-fac9-466b-941b-590b2a401aa6"
       },
-      "outputs": [],
+      "outputs": [
+        {
+          "name": "stderr",
+          "output_type": "stream",
+          "text": [
+            "huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\n",
+            "To disable this warning, you can either:\n",
+            "\t- Avoid using `tokenizers` before the fork if possible\n",
+            "\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\n"
+          ]
+        }
+      ],
       "source": [
         "!pip install markdownify duckduckgo-search spaces gradio-tools langchain langchain-community langchain-huggingface faiss-cpu --upgrade -q"
       ]
@@ -150,7 +172,7 @@
     },
     {
       "cell_type": "code",
-      "execution_count": null,
+      "execution_count": 13,
       "metadata": {
         "colab": {
           "base_uri": "https://localhost:8080/"
@@ -160,10 +182,10 @@
       },
       "outputs": [],
       "source": [
-        "from transformers.agents import HfApiEngine\n",
+        "from smolagents import HfApiModel\n",
         "\n",
-        "llm_model = \"Qwen/Qwen2.5-72B-Instruct\"\n",
-        "llm_engine = HfApiEngine(llm_model)"
+        "model_id = \"Qwen/Qwen2.5-72B-Instruct\"\n",
+        "model = HfApiModel(model_id)"
       ]
     },
     {
@@ -203,9 +225,9 @@
       "source": [
         "### 2.1.1 Build our multi-tool web agent 🤖\n",
         "\n",
-        "Now that we've set up the basic search and webpage tools, let's build our **multi-tool web agent**. This agent will combine several tools to perform more complex tasks, leveraging the capabilities of the `ReactJsonAgent`.\n",
+        "Now that we've set up the basic search and webpage tools, let's build our **multi-tool web agent**. This agent will combine several tools to perform more complex tasks, leveraging the capabilities of the `ToolCallingAgent`.\n",
         "\n",
-        "The `ReactJsonAgent` is particularly well-suited for web search tasks because its JSON action formulation requires only simple arguments and works seamlessly in sequential chains of single actions. This makes it an excellent choice for scenarios where we need to search the web for relevant information and retrieve detailed content from specific web pages. In contrast, `CodeAgent` action formulation is better suited for scenarios involving numerous or parallel tool calls.\n",
+        "The `ToolCallingAgent` is particularly well-suited for web search tasks because its JSON action formulation requires only simple arguments and works seamlessly in sequential chains of single actions. This makes it an excellent choice for scenarios where we need to search the web for relevant information and retrieve detailed content from specific web pages. In contrast, `CodeAgent` action formulation is better suited for scenarios involving numerous or parallel tool calls.\n",
         "\n",
         "By integrating multiple tools, we can ensure that our agent interacts with the web in a sophisticated and efficient manner.\n",
         "\n",
@@ -215,18 +237,17 @@
     },
     {
       "cell_type": "code",
-      "execution_count": 5,
+      "execution_count": 14,
       "metadata": {
         "id": "NVnfGll5B0Ma"
       },
       "outputs": [],
       "source": [
-        "from transformers.agents import ReactCodeAgent, ReactJsonAgent, ManagedAgent\n",
-        "from transformers.agents.search import DuckDuckGoSearchTool, VisitWebpageTool\n",
+        "from smolagents import CodeAgent, ToolCallingAgent, ManagedAgent, DuckDuckGoSearchTool, VisitWebpageTool\n",
         "\n",
-        "web_agent = ReactJsonAgent(\n",
+        "web_agent = ToolCallingAgent(\n",
         "    tools=[DuckDuckGoSearchTool(), VisitWebpageTool()],\n",
-        "    llm_engine=llm_engine\n",
+        "    model=model\n",
         ")"
       ]
     },
@@ -241,15 +262,15 @@
     },
     {
       "cell_type": "code",
-      "execution_count": 6,
+      "execution_count": 15,
       "metadata": {
         "id": "igUZsVP3B0Mb"
       },
       "outputs": [],
       "source": [
         "managed_web_agent = ManagedAgent(\n",
         "    agent=web_agent,\n",
-        "    name=\"search\",\n",
+        "    name=\"search_agent\",\n",
         "    description=\"Runs web searches for you. Give it your query as an argument.\",\n",
         ")"
       ]
@@ -296,7 +317,7 @@
     },
     {
       "cell_type": "code",
-      "execution_count": null,
+      "execution_count": 16,
       "metadata": {
         "colab": {
           "base_uri": "https://localhost:8080/",
@@ -682,8 +703,8 @@
       },
       "outputs": [],
       "source": [
-        "retriever_agent = ReactJsonAgent(\n",
-        "    tools=[huggingface_doc_retriever_tool, peft_issues_retriever_tool], llm_engine=llm_engine, max_iterations=4, verbose=2\n",
+        "retriever_agent = ToolCallingAgent(\n",
+        "    tools=[huggingface_doc_retriever_tool, peft_issues_retriever_tool], model=model, max_iterations=4, verbose=2\n",
         ")"
       ]
     },
@@ -697,7 +718,7 @@
       "source": [
         "managed_retriever_agent = ManagedAgent(\n",
         "    agent=retriever_agent,\n",
-        "    name=\"retriever\",\n",
+        "    name=\"retriever_agent\",\n",
         "    description=\"Retrieves documents from the knowledge base for you that are close to the input query. Give it your query as an argument. The knowledge base includes Hugging Face documentation and PEFT issues.\",\n",
         ")"
       ]
@@ -773,8 +794,8 @@
       },
       "outputs": [],
       "source": [
-        "image_generation_tool = load_tool(\"m-ric/text-to-image\", cache=False)\n",
-        "image_generation_agent = CodeAgent(tools=[prompt_generator_tool, image_generation_tool], llm_engine=llm_engine)"
+        "image_generation_tool = load_tool(\"m-ric/text-to-image\", trust_remote_code=True)\n",
+        "image_generation_agent = CodeAgent(tools=[prompt_generator_tool, image_generation_tool], model=model)"
       ]
     },
     {
@@ -796,7 +817,7 @@
       "source": [
         "managed_image_generation_agent = ManagedAgent(\n",
         "    agent=image_generation_agent,\n",
-        "    name=\"image_generation\",\n",
+        "    name=\"image_generation_agent\",\n",
         "    description=\"Generates images from text prompts. Give it your prompt as an argument.\",\n",
         "    additional_prompting=\"\\n\\nYour final answer MUST BE only the generated image location.\"\n",
         ")"
@@ -828,9 +849,9 @@
       },
       "outputs": [],
       "source": [
-        "manager_agent = ReactCodeAgent(\n",
+        "manager_agent = CodeAgent(\n",
         "    tools=[],\n",
-        "    llm_engine=llm_engine,\n",
+        "    model=model,\n",
         "    managed_agents=[managed_web_agent, managed_retriever_agent, managed_image_generation_agent],\n",
         "    additional_authorized_imports=[\"time\", \"datetime\", \"PIL\"],\n",
         ")"
@@ -1551,9 +1572,9 @@
       "provenance": []
     },
     "kernelspec": {
-      "display_name": "cookbook2",
+      "display_name": "test2",
       "language": "python",
-      "name": "cookbook2"
+      "name": "test2"
     },
     "language_info": {
       "codemirror_mode": {

diff --git a/notebooks/zh-CN/_toctree.yml b/notebooks/zh-CN/_toctree.yml
@@ -72,9 +72,7 @@
         - local: agents
           title: 使用 Transformers Agents 构建具有工具调用超能力的智能体
         - local: agent_rag
-          title: 智能体 RAG 通过查询重构和自查询来增强你的 RAG 
-        - local: agent_change_llm
-          title: 从任意的 LLM 推理提供商中创建一个 Transformers 智能体
+          title: 智能体 RAG 通过查询重构和自查询来增强你的 RAG
 
 - title: 企业级使用指南 (Cookbook)
   isExpanded: True