Upload 56 files

.gitattributes

@@ -1,35 +1,34 @@
-*.7z filter=lfs diff=lfs merge=lfs -text
-*.arrow filter=lfs diff=lfs merge=lfs -text
-*.bin filter=lfs diff=lfs merge=lfs -text
-*.bz2 filter=lfs diff=lfs merge=lfs -text
-*.ckpt filter=lfs diff=lfs merge=lfs -text
-*.ftz filter=lfs diff=lfs merge=lfs -text
-*.gz filter=lfs diff=lfs merge=lfs -text
-*.h5 filter=lfs diff=lfs merge=lfs -text
-*.joblib filter=lfs diff=lfs merge=lfs -text
-*.lfs.* filter=lfs diff=lfs merge=lfs -text
-*.mlmodel filter=lfs diff=lfs merge=lfs -text
-*.model filter=lfs diff=lfs merge=lfs -text
-*.msgpack filter=lfs diff=lfs merge=lfs -text
-*.npy filter=lfs diff=lfs merge=lfs -text
-*.npz filter=lfs diff=lfs merge=lfs -text
-*.onnx filter=lfs diff=lfs merge=lfs -text
-*.ot filter=lfs diff=lfs merge=lfs -text
-*.parquet filter=lfs diff=lfs merge=lfs -text
-*.pb filter=lfs diff=lfs merge=lfs -text
-*.pickle filter=lfs diff=lfs merge=lfs -text
-*.pkl filter=lfs diff=lfs merge=lfs -text
-*.pt filter=lfs diff=lfs merge=lfs -text
-*.pth filter=lfs diff=lfs merge=lfs -text
-*.rar filter=lfs diff=lfs merge=lfs -text
-*.safetensors filter=lfs diff=lfs merge=lfs -text
-saved_model/**/* filter=lfs diff=lfs merge=lfs -text
-*.tar.* filter=lfs diff=lfs merge=lfs -text
-*.tar filter=lfs diff=lfs merge=lfs -text
-*.tflite filter=lfs diff=lfs merge=lfs -text
-*.tgz filter=lfs diff=lfs merge=lfs -text
-*.wasm filter=lfs diff=lfs merge=lfs -text
-*.xz filter=lfs diff=lfs merge=lfs -text
-*.zip filter=lfs diff=lfs merge=lfs -text
-*.zst filter=lfs diff=lfs merge=lfs -text
-*tfevents* filter=lfs diff=lfs merge=lfs -text
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,10 +1,14 @@
----
-title: Devmodetest2
-emoji: 🐨
-colorFrom: blue
-colorTo: red
-sdk: docker
-pinned: false
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+---
+title: InfiniInference
+emoji: 💻🐳
+colorFrom: gray
+colorTo: green
+sdk: docker
+pinned: false
+tags:
+- jupyterlab
+suggested_storage: small
+license: other
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

experimental.ipynb

@@ -0,0 +1,64 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "initial_id",
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "%pip install -r requirements.txt"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "outputs": [],
+   "source": [
+    "%env LANGCHAIN_TRACING_V2=true\n",
+    "%env LANGCHAIN_ENDPOINT=\"https://api.smith.langchain.com\"\n",
+    "%env LANGCHAIN_API_KEY=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\n",
+    "%env LANGCHAIN_PROJECT=TestProject"
+   ],
+   "metadata": {
+    "collapsed": false
+   },
+   "id": "66a5af01e7e46ad9"
+  },
+  {
+   "cell_type": "code",
+   "outputs": [],
+   "source": [
+    "from infiniInference.handler import start\n",
+    "from infiniInference.agent_factory import create_agent\n",
+    "start()"
+   ],
+   "metadata": {
+    "collapsed": false
+   },
+   "id": "55569c1ec7bfe3f6"
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 2
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython2",
+   "version": "2.7.6"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

infiniInference/agent_factory.py

@@ -0,0 +1,20 @@
+from langchain.agents import AgentExecutor
+from transformers import PreTrainedModel
+
+
+def create_agent(llm: PreTrainedModel, tools: list, system_prompt: str):
+    # Each worker node will be given a name and some tools.
+    prompt = ChatPromptTemplate.from_messages(
+        [
+            (
+                "system",
+                system_prompt,
+            ),
+            MessagesPlaceholder(variable_name="messages"),
+            MessagesPlaceholder(variable_name="agent_scratchpad"),
+        ]
+    )
+    agent = prompt | llm.bind_tools(tools)
+    # create_openai_tools_agent(llm, tools, prompt)
+    executor = AgentExecutor(agent=agent, tools=tools)
+    return executor

infiniInference/agent_node.py

@@ -0,0 +1,4 @@
+from langchain_core.messages import BaseMessage, HumanMessage
+def agent_node(state, agent, name):
+    result = agent.invoke(state)
+    return {"messages": [HumanMessage(content=result["output"], name=name)]}

infiniInference/handler.py

@@ -0,0 +1,48 @@
+import getpass
+import os
+from langchain_core.messages import BaseMessage
+from infiniInference.agent_factory import create_agent
+from infiniInference.supervisor import llm
+
+
+def _set_if_undefined(var: str):
+    if not os.environ.get(var):
+        os.environ[var] = getpass.getpass(f"Please provide your {var}")
+
+
+#_set_if_undefined("OPENAI_API_KEY")
+# Optional, add tracing in LangSmith
+os.environ["LANGCHAIN_TRACING_V2"] = "true"
+os.environ["LANGCHAIN_PROJECT"] = "Agent test"
+
+import operator
+from typing import Annotated, Any, Dict, List, Optional, Sequence, TypedDict
+import functools
+
+from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
+from langgraph.graph import StateGraph, END
+
+
+# The agent state is the input to each node in the graph
+class AgentState(TypedDict):
+    # The annotation tells the graph that new messages will always
+    # be added to the current states
+    messages: Annotated[Sequence[BaseMessage], operator.add]
+    # The 'next' field indicates where to route to next
+    next: str
+
+research_agent = create_agent(llm, [tavily_tool], "You are a web researcher.")
+research_node = functools.partial(agent_node, agent=research_agent, name="Researcher")
+
+# NOTE: THIS PERFORMS ARBITRARY CODE EXECUTION. PROCEED WITH CAUTION
+code_agent = create_agent(
+    llm,
+    [python_repl_tool],
+    "You may generate safe python code to analyze data and generate charts using matplotlib.",
+)
+code_node = functools.partial(agent_node, agent=code_agent, name="Coder")
+
+workflow = StateGraph(AgentState)
+workflow.add_node("Researcher", research_node)
+workflow.add_node("Coder", code_node)
+workflow.add_node("supervisor", supervisor_chain)

infiniInference/supervisor.py

@@ -0,0 +1,97 @@
+from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
+# from langchain_core.output_parsers.openai_functions import JsonOutputFunctionsParser
+import torch
+from transformers import AutoTokenizer, AutoModelForSequenceClassification, AutoModelForCausalLM, pipeline
+from langchain_huggingface import HuggingFacePipeline
+import getpass
+from langchain.chains import SimpleChain
+from langchain_mistralai import ChatMistralAI
+token = getpass.getpass("Token: ")
+
+members = ["Researcher", "Coder"]
+system_prompt = (
+    "You are a supervisor tasked with managing a conversation between the"
+    " following workers:  {members}. Given the following user request,"
+    " respond with the worker to act next. Each worker will perform a"
+    " task and respond with their results and status. When finished,"
+    " respond with FINISH."
+)
+# Our team supervisor is an LLM node. It just picks the next agent to process
+# and decides when the work is completed
+options = ["FINISH"] + members
+# Using openai function calling can make output parsing easier for us
+function_def = {
+    "name": "route",
+    "description": "Select the next role.",
+    "parameters": {
+        "title": "routeSchema",
+        "type": "object",
+        "properties": {
+            "next": {
+                "title": "Next",
+                "anyOf": [
+                    {"enum": options},
+                ],
+            }
+        },
+        "required": ["next"],
+    },
+}
+prompt = ChatPromptTemplate.from_messages(
+    [
+        ("system", system_prompt),
+        MessagesPlaceholder(variable_name="messages"),
+        (
+            "system",
+            "Given the conversation above, who should act next?"
+            " Or should we FINISH? Select one of: {options}",
+        ),
+    ]
+).partial(options=str(options), members=", ".join(members))
+path = "mistralai/Mistral-7B-Instruct-v0.3"
+
+model = AutoModelForCausalLM.from_pretrained(
+    path,
+    torch_dtype=torch.float16,
+    low_cpu_mem_usage=True,
+    trust_remote_code=True,
+    device_map="auto",
+    token=token
+)
+
+tokenizer = AutoTokenizer.from_pretrained(path, token=token)
+tokenizer.add_special_tokens({"pad_token": "[PAD]"})
+tokenizer.padding_side = "left"
+
+pipe = pipeline(task='text-generation', model=model, tokenizer=tokenizer,
+                num_return_sequences=1,
+                eos_token_id=tokenizer.eos_token_id, pad_token_id=tokenizer.eos_token_id,
+                max_new_tokens=260, temperature=0.7, do_sample=True)
+
+llm = HuggingFacePipeline(pipeline=pipe)
+
+
+
+def custom_function(input_text):
+    # Example function logic
+    return {"output": "processed " + input_text}
+
+from langchain.tools import Tool
+
+class MyCustomTool(Tool):
+    def call(self, input_text: str) -> str:
+        # Custom tool logic here
+        return f"Processed: {input_text}"
+
+# Initialize the custom tool
+my_tool = MyCustomTool()
+
+chain = prompt | llm | [my_tool]
+
+# Define the input text
+input_text = "Your input text here"
+
+# Run the chain with the input text
+result = chain.invoke(input_text)
+
+print(result)

infiniInference/testme.ipynb

@@ -0,0 +1,35 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "initial_id",
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": "test    "
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 2
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython2",
+   "version": "2.7.6"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

myapp.py

@@ -0,0 +1 @@
+print("Hello remote")

perm/Agents9 (2).ipynb

@@ -0,0 +1,552 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "124c8be0-1c20-4ea1-aeee-106d3fb7143f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from langchain_huggingface import HuggingFacePipeline, ChatHuggingFace\n",
+    "from transformers import AutoTokenizer, AutoModelForSequenceClassification, AutoModelForCausalLM, pipeline\n",
+    "import torch\n",
+    "from langchain_core.messages import (\n",
+    "    HumanMessage,\n",
+    "    SystemMessage,\n",
+    "    BaseMessage\n",
+    ")\n",
+    "from typing import (\n",
+    "    Any,\n",
+    "    Callable,\n",
+    "    Dict,\n",
+    "    List,\n",
+    "    Literal,\n",
+    "    Optional,\n",
+    "    Sequence,\n",
+    "    Type,\n",
+    "    Union,\n",
+    "    cast,\n",
+    ")\n",
+    "\n",
+    "class HuggingFaceAI(ChatHuggingFace):\n",
+    "    \n",
+    "    def _to_chat_prompt(\n",
+    "        self,\n",
+    "        messages: List[BaseMessage],\n",
+    "    ) -> str:\n",
+    "        \"\"\"Convert a list of messages into a prompt format expected by wrapped LLM.\"\"\"\n",
+    "        if not messages:\n",
+    "            raise ValueError(\"At least one HumanMessage must be provided!\")\n",
+    "\n",
+    "        if not isinstance(messages[-1], HumanMessage) and not isinstance(messages[-1], SystemMessage) :\n",
+    "            raise ValueError(\"Last message must be a HumanMessage or SystemMessage!!!\")\n",
+    "\n",
+    "        messages_dicts = [self._to_chatml_format(m) for m in messages]\n",
+    "\n",
+    "        return self.tokenizer.apply_chat_template(\n",
+    "            messages_dicts, tokenize=False, add_generation_prompt=True\n",
+    "        )"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "3ab4c82b-4268-4ffa-b814-967cf5e347d2",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "bb1698c756e946beb49a2afcfde2e51e",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "tokenizer_config.json:   0%|          | 0.00/137k [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "2fe3d3847e4e4cfa8b124211c679597c",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "tokenizer.model:   0%|          | 0.00/587k [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "8dc27068bab14901a3595e921a34c4ef",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "tokenizer.json:   0%|          | 0.00/1.96M [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "bc109fc9216449f19e610b30a9879b27",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "special_tokens_map.json:   0%|          | 0.00/414 [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "c3180e986f4d411d8511e361e4f3cd8f",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "config.json:   0%|          | 0.00/601 [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4c3c52d9f0c04b1eb3eb0c5ebd3e14c8",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "model.safetensors.index.json:   0%|          | 0.00/23.9k [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "824ba11de3ba4cc7aa713779f3971025",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Downloading shards:   0%|          | 0/3 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "8c9b9325e7584ccf93733ab371493b1a",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "model-00001-of-00003.safetensors:   0%|          | 0.00/4.95G [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "2a5cfc651b7f4b6793305c507bb4f1ca",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "model-00002-of-00003.safetensors:   0%|          | 0.00/5.00G [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "1aefc575e0c34cea950ca2e660024cc2",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "model-00003-of-00003.safetensors:   0%|          | 0.00/4.55G [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "511d965c616345eaaa447a3f6ae3d8b8",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Loading checkpoint shards:   0%|          | 0/3 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "a222b677f36e430383e9470d87a143c0",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "generation_config.json:   0%|          | 0.00/116 [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "AIMessage(content='<s>[INST] Hugging Face is [/INST] Hugging Face is a technology company that specializes in natural language processing (NLP). They are best known for their transformers library, which is a state-of-the-art machine learning framework for NLP tasks. The transformers library includes pre-trained models for a wide range of NLP tasks such as language translation, text classification, and question answering. Hugging Face also provides a platform for training, sharing, and using NLP models, called the Hugging Face Model Hub', id='run-9b0b07cc-2121-4c78-8ebf-61df58694193-0')"
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "from langchain_huggingface import ChatHuggingFace, HuggingFacePipeline\n",
+    "\n",
+    "llm = HuggingFacePipeline.from_model_id(\n",
+    "    model_id=\"mistralai/Mistral-7B-Instruct-v0.3\",\n",
+    "    device_map=\"auto\",\n",
+    "    task=\"text-generation\",\n",
+    "    pipeline_kwargs={\n",
+    "        \"max_new_tokens\": 100,\n",
+    "        \"top_k\": 50,\n",
+    "    },\n",
+    ")\n",
+    "\n",
+    "llm_engine_hf = HuggingFaceAI(llm=llm)\n",
+    "llm_engine_hf.invoke(\"Hugging Face is\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "8cf48c8f-27c2-4c22-a219-965ad3621bc2",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "from langchain.agents import AgentExecutor, create_openai_tools_agent\n",
+    "from langchain_core.messages import BaseMessage, HumanMessage, SystemMessage, AIMessage\n",
+    "from langchain_openai import ChatOpenAI\n",
+    "\n",
+    "\n",
+    "def create_agent(llm: ChatHuggingFace, tools: list, system_prompt: str):\n",
+    "    # Each worker node will be given a name and some tools.\n",
+    "    prompt = ChatPromptTemplate.from_messages(\n",
+    "        [\n",
+    "            (\n",
+    "                \"system\",\n",
+    "                system_prompt,\n",
+    "            ),\n",
+    "            MessagesPlaceholder(variable_name=\"messages\"),\n",
+    "            MessagesPlaceholder(variable_name=\"agent_scratchpad\"),\n",
+    "        ]\n",
+    "    )\n",
+    "    agent = create_openai_tools_agent(llm, tools, prompt)\n",
+    "    executor = AgentExecutor(agent=agent, tools=tools)\n",
+    "    return executor"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "b2968526-9c48-45fe-b783-707873af7436",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "from typing import Annotated, List, Tuple, Union\n",
+    "\n",
+    "from langchain_core.tools import tool\n",
+    "from langchain_experimental.tools import PythonREPLTool\n",
+    "\n",
+    "\n",
+    "# This executes code locally, which can be unsafe\n",
+    "python_repl_tool = PythonREPLTool()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "fd6d6e2f-d327-4a03-8693-9b6e20170826",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "chat_model = llm_engine_hf"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "5cf6d454-81fb-43cd-a472-519df681ca32",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "def agent_node(state, agent, name):\n",
+    "    result = agent.invoke(state)\n",
+    "    return {\"messages\": [HumanMessage(content=result[\"output\"], name=name)]}"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "d35ca09b-1310-466d-924a-31f761a1f700",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from langchain_core.tools import tool\n",
+    "\n",
+    "@tool\n",
+    "def multiply(a: int, b: int) -> int:\n",
+    "    \"\"\"Multiply two numbers.\"\"\"\n",
+    "    print(\"Multiply used\")\n",
+    "    return a * b"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "id": "1579a0b5-bc37-45d4-aba4-244c3891f0f6",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "from langchain_core.output_parsers import StrOutputParser\n",
+    "from langchain_core.prompts import PromptTemplate, ChatPromptTemplate, MessagesPlaceholder\n",
+    "from langchain_core.runnables import RunnableLambda    \n",
+    "\n",
+    "def stripPrompt(info):\n",
+    "    print(info)\n",
+    "    eot_token = \"[/INST] \"\n",
+    "    i = info.content.rfind(eot_token)\n",
+    "    if i == -1:\n",
+    "        return info\n",
+    "    \n",
+    "    info.content = info.content[i + len(eot_token):]\n",
+    "    \n",
+    "    return info.content\n",
+    "   \n",
+    "members = [\"Multiplier\", \"Coder\"]\n",
+    "# Our team supervisor is an LLM node. It just picks the next agent to process\n",
+    "# and decides when the work is completed\n",
+    "options = [\"FINISH\"] + members\n",
+    "system_prompt = (\n",
+    "    \"You are a supervisor tasked with managing a conversation between the\"\n",
+    "    \" following workers:  {members}. Given the following user request,\"\n",
+    "    \" respond with the worker to act next. Each worker will perform a\"\n",
+    "    \" task and respond with their results and status. When finished,\"\n",
+    "    \" respond with FINISH.\"\n",
+    ")\n",
+    "\n",
+    "\n",
+    "prompt = ChatPromptTemplate.from_messages(\n",
+    "    [\n",
+    "        (\"human\", system_prompt),\n",
+    "        (\"assistant\", \"ok\"),\n",
+    "        MessagesPlaceholder(variable_name=\"messages\"),\n",
+    "        (\"assistant\", \"ok\"),\n",
+    "        (\n",
+    "            \"human\",\n",
+    "            \"Given the conversation above, who should act next?\"\n",
+    "            \" Or should we FINISH? Select one of: {options}\",\n",
+    "        ),\n",
+    "    ]\n",
+    ").partial(options=str(options), members=\", \".join(members))\n",
+    "\n",
+    "chain = ( prompt | chat_model | RunnableLambda(stripPrompt))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "id": "dbe4efe7-220a-42f9-badc-0e27329c8225",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "import operator\n",
+    "from typing import Annotated, Any, Dict, List, Optional, Sequence, TypedDict\n",
+    "import functools\n",
+    "\n",
+    "from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder\n",
+    "from langgraph.graph import StateGraph, END\n",
+    "\n",
+    "\n",
+    "\n",
+    "\n",
+    "\n",
+    "multiply_agent = create_agent(chat_model, [multiply], \"You multiply numbers\")\n",
+    "multiply_node = functools.partial(agent_node, agent=multiply_agent, name=\"Multiplier\")\n",
+    "#research_node = functools.partial(agent_node, agent=research_agent, name=\"Researcher\")\n",
+    "\n",
+    "# NOTE: THIS PERFORMS ARBITRARY CODE EXECUTION. PROCEED WITH CAUTION\n",
+    "code_agent = create_agent(\n",
+    "    chat_model,\n",
+    "    [python_repl_tool],\n",
+    "    \"You may generate safe python code to analyze data and generate charts using matplotlib.\",\n",
+    ")\n",
+    "code_node = functools.partial(agent_node, agent=code_agent, name=\"Coder\")\n",
+    "\n",
+    "workflow = StateGraph(AgentState)\n",
+    "workflow.add_node(\"Multiplier\", multiply_node)\n",
+    "workflow.add_node(\"Coder\", code_node)\n",
+    "workflow.add_node(\"supervisor\", chain)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "id": "74714d27-60b1-4ae3-9993-d0c10b77e417",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "for member in members:\n",
+    "    # We want our workers to ALWAYS \"report back\" to the supervisor when done\n",
+    "    workflow.add_edge(member, \"supervisor\")\n",
+    "# The supervisor populates the \"next\" field in the graph state\n",
+    "# which routes to a node or finishes\n",
+    "conditional_map = {k: k for k in members}\n",
+    "conditional_map[\"FINISH\"] = END\n",
+    "workflow.add_conditional_edges(\"supervisor\", lambda x: x[\"next\"], conditional_map)\n",
+    "# Finally, add entrypoint\n",
+    "workflow.set_entry_point(\"supervisor\")\n",
+    "\n",
+    "graph = workflow.compile()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 25,
+   "id": "e6dc680d-15ec-484b-9a72-dc8af0f04234",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "content=\"<s>[INST] You are a supervisor tasked with managing a conversation between the following workers:  Multiplier, Coder. Given the following user request, respond with the worker to act next. Each worker will perform a task and respond with their results and status. When finished, respond with FINISH. [/INST]ok</s>[INST] What is 4 multiplied by 35 [/INST]ok</s>[INST] Given the conversation above, who should act next? Or should we FINISH? Select one of: ['FINISH', 'Multiplier', 'Coder'] [/INST] Multiplier (since they provided the result of the multiplication)\" id='run-8c0df905-9902-4ac2-8327-f4e4fb612cc2-0'\n"
+     ]
+    },
+    {
+     "ename": "InvalidUpdateError",
+     "evalue": "Expected dict, got Multiplier (since they provided the result of the multiplication)",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mInvalidUpdateError\u001b[0m                        Traceback (most recent call last)",
+      "Cell \u001b[0;32mIn[25], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m s \u001b[38;5;129;01min\u001b[39;00m graph\u001b[38;5;241m.\u001b[39mstream(\n\u001b[1;32m      2\u001b[0m     {\n\u001b[1;32m      3\u001b[0m         \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mmessages\u001b[39m\u001b[38;5;124m\"\u001b[39m: [\n\u001b[1;32m      4\u001b[0m             HumanMessage(content\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mWhat is 4 multiplied by 35\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[1;32m      5\u001b[0m         ]\n\u001b[1;32m      6\u001b[0m     }\n\u001b[1;32m      7\u001b[0m ):\n\u001b[1;32m      8\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m__end__\u001b[39m\u001b[38;5;124m\"\u001b[39m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;129;01min\u001b[39;00m s:\n\u001b[1;32m      9\u001b[0m         \u001b[38;5;28mprint\u001b[39m(s)\n",
+      "File \u001b[0;32m~/miniconda/lib/python3.9/site-packages/langgraph/pregel/__init__.py:963\u001b[0m, in \u001b[0;36mPregel.stream\u001b[0;34m(self, input, config, stream_mode, output_keys, input_keys, interrupt_before, interrupt_after, debug)\u001b[0m\n\u001b[1;32m    960\u001b[0m         \u001b[38;5;28;01mdel\u001b[39;00m fut, task\n\u001b[1;32m    962\u001b[0m \u001b[38;5;66;03m# panic on failure or timeout\u001b[39;00m\n\u001b[0;32m--> 963\u001b[0m \u001b[43m_panic_or_proceed\u001b[49m\u001b[43m(\u001b[49m\u001b[43mdone\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43minflight\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mstep\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    964\u001b[0m \u001b[38;5;66;03m# don't keep futures around in memory longer than needed\u001b[39;00m\n\u001b[1;32m    965\u001b[0m \u001b[38;5;28;01mdel\u001b[39;00m done, inflight, futures\n",
+      "File \u001b[0;32m~/miniconda/lib/python3.9/site-packages/langgraph/pregel/__init__.py:1489\u001b[0m, in \u001b[0;36m_panic_or_proceed\u001b[0;34m(done, inflight, step)\u001b[0m\n\u001b[1;32m   1487\u001b[0m             inflight\u001b[38;5;241m.\u001b[39mpop()\u001b[38;5;241m.\u001b[39mcancel()\n\u001b[1;32m   1488\u001b[0m         \u001b[38;5;66;03m# raise the exception\u001b[39;00m\n\u001b[0;32m-> 1489\u001b[0m         \u001b[38;5;28;01mraise\u001b[39;00m exc\n\u001b[1;32m   1491\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m inflight:\n\u001b[1;32m   1492\u001b[0m     \u001b[38;5;66;03m# if we got here means we timed out\u001b[39;00m\n\u001b[1;32m   1493\u001b[0m     \u001b[38;5;28;01mwhile\u001b[39;00m inflight:\n\u001b[1;32m   1494\u001b[0m         \u001b[38;5;66;03m# cancel all pending tasks\u001b[39;00m\n",
+      "File \u001b[0;32m~/miniconda/lib/python3.9/concurrent/futures/thread.py:52\u001b[0m, in \u001b[0;36m_WorkItem.run\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m     49\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m\n\u001b[1;32m     51\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m---> 52\u001b[0m     result \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfn\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     53\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mBaseException\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m exc:\n\u001b[1;32m     54\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mfuture\u001b[38;5;241m.\u001b[39mset_exception(exc)\n",
+      "File \u001b[0;32m~/miniconda/lib/python3.9/site-packages/langgraph/pregel/retry.py:66\u001b[0m, in \u001b[0;36mrun_with_retry\u001b[0;34m(task, retry_policy)\u001b[0m\n\u001b[1;32m     64\u001b[0m task\u001b[38;5;241m.\u001b[39mwrites\u001b[38;5;241m.\u001b[39mclear()\n\u001b[1;32m     65\u001b[0m \u001b[38;5;66;03m# run the task\u001b[39;00m\n\u001b[0;32m---> 66\u001b[0m \u001b[43mtask\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mproc\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43minvoke\u001b[49m\u001b[43m(\u001b[49m\u001b[43mtask\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43minput\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mtask\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mconfig\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     67\u001b[0m \u001b[38;5;66;03m# if successful, end\u001b[39;00m\n\u001b[1;32m     68\u001b[0m \u001b[38;5;28;01mbreak\u001b[39;00m\n",
+      "File \u001b[0;32m~/miniconda/lib/python3.9/site-packages/langchain_core/runnables/base.py:2495\u001b[0m, in \u001b[0;36mRunnableSequence.invoke\u001b[0;34m(self, input, config, **kwargs)\u001b[0m\n\u001b[1;32m   2493\u001b[0m             \u001b[38;5;28minput\u001b[39m \u001b[38;5;241m=\u001b[39m step\u001b[38;5;241m.\u001b[39minvoke(\u001b[38;5;28minput\u001b[39m, config, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n\u001b[1;32m   2494\u001b[0m         \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m-> 2495\u001b[0m             \u001b[38;5;28minput\u001b[39m \u001b[38;5;241m=\u001b[39m \u001b[43mstep\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43minvoke\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43minput\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mconfig\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   2496\u001b[0m \u001b[38;5;66;03m# finish the root run\u001b[39;00m\n\u001b[1;32m   2497\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mBaseException\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m e:\n",
+      "File \u001b[0;32m~/miniconda/lib/python3.9/site-packages/langgraph/utils.py:86\u001b[0m, in \u001b[0;36mRunnableCallable.invoke\u001b[0;34m(self, input, config, **kwargs)\u001b[0m\n\u001b[1;32m     84\u001b[0m kwargs \u001b[38;5;241m=\u001b[39m {\u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mkwargs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs}\n\u001b[1;32m     85\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mtrace:\n\u001b[0;32m---> 86\u001b[0m     ret \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_call_with_config\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m     87\u001b[0m \u001b[43m        \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfunc\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43minput\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mmerge_configs\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mconfig\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mconfig\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\n\u001b[1;32m     88\u001b[0m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     89\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m     90\u001b[0m     config \u001b[38;5;241m=\u001b[39m merge_configs(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mconfig, config)\n",
+      "File \u001b[0;32m~/miniconda/lib/python3.9/site-packages/langchain_core/runnables/base.py:1596\u001b[0m, in \u001b[0;36mRunnable._call_with_config\u001b[0;34m(self, func, input, config, run_type, **kwargs)\u001b[0m\n\u001b[1;32m   1592\u001b[0m     context \u001b[38;5;241m=\u001b[39m copy_context()\n\u001b[1;32m   1593\u001b[0m     context\u001b[38;5;241m.\u001b[39mrun(_set_config_context, child_config)\n\u001b[1;32m   1594\u001b[0m     output \u001b[38;5;241m=\u001b[39m cast(\n\u001b[1;32m   1595\u001b[0m         Output,\n\u001b[0;32m-> 1596\u001b[0m         \u001b[43mcontext\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mrun\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   1597\u001b[0m \u001b[43m            \u001b[49m\u001b[43mcall_func_with_variable_args\u001b[49m\u001b[43m,\u001b[49m\u001b[43m  \u001b[49m\u001b[38;5;66;43;03m# type: ignore[arg-type]\u001b[39;49;00m\n\u001b[1;32m   1598\u001b[0m \u001b[43m            \u001b[49m\u001b[43mfunc\u001b[49m\u001b[43m,\u001b[49m\u001b[43m  \u001b[49m\u001b[38;5;66;43;03m# type: ignore[arg-type]\u001b[39;49;00m\n\u001b[1;32m   1599\u001b[0m \u001b[43m            \u001b[49m\u001b[38;5;28;43minput\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m  \u001b[49m\u001b[38;5;66;43;03m# type: ignore[arg-type]\u001b[39;49;00m\n\u001b[1;32m   1600\u001b[0m \u001b[43m            \u001b[49m\u001b[43mconfig\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1601\u001b[0m \u001b[43m            \u001b[49m\u001b[43mrun_manager\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1602\u001b[0m \u001b[43m            \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1603\u001b[0m \u001b[43m        \u001b[49m\u001b[43m)\u001b[49m,\n\u001b[1;32m   1604\u001b[0m     )\n\u001b[1;32m   1605\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mBaseException\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[1;32m   1606\u001b[0m     run_manager\u001b[38;5;241m.\u001b[39mon_chain_error(e)\n",
+      "File \u001b[0;32m~/miniconda/lib/python3.9/site-packages/langchain_core/runnables/config.py:380\u001b[0m, in \u001b[0;36mcall_func_with_variable_args\u001b[0;34m(func, input, config, run_manager, **kwargs)\u001b[0m\n\u001b[1;32m    378\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m run_manager \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;129;01mand\u001b[39;00m accepts_run_manager(func):\n\u001b[1;32m    379\u001b[0m     kwargs[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mrun_manager\u001b[39m\u001b[38;5;124m\"\u001b[39m] \u001b[38;5;241m=\u001b[39m run_manager\n\u001b[0;32m--> 380\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mfunc\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43minput\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/miniconda/lib/python3.9/site-packages/langgraph/pregel/write.py:97\u001b[0m, in \u001b[0;36mChannelWrite._write\u001b[0;34m(self, input, config)\u001b[0m\n\u001b[1;32m     93\u001b[0m \u001b[38;5;66;03m# process entries into values\u001b[39;00m\n\u001b[1;32m     94\u001b[0m values \u001b[38;5;241m=\u001b[39m [\n\u001b[1;32m     95\u001b[0m     \u001b[38;5;28minput\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m write\u001b[38;5;241m.\u001b[39mvalue \u001b[38;5;129;01mis\u001b[39;00m PASSTHROUGH \u001b[38;5;28;01melse\u001b[39;00m write\u001b[38;5;241m.\u001b[39mvalue \u001b[38;5;28;01mfor\u001b[39;00m write \u001b[38;5;129;01min\u001b[39;00m entries\n\u001b[1;32m     96\u001b[0m ]\n\u001b[0;32m---> 97\u001b[0m values \u001b[38;5;241m=\u001b[39m [\n\u001b[1;32m     98\u001b[0m     val \u001b[38;5;28;01mif\u001b[39;00m write\u001b[38;5;241m.\u001b[39mmapper \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;28;01melse\u001b[39;00m write\u001b[38;5;241m.\u001b[39mmapper\u001b[38;5;241m.\u001b[39minvoke(val, config)\n\u001b[1;32m     99\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m val, write \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mzip\u001b[39m(values, entries)\n\u001b[1;32m    100\u001b[0m ]\n\u001b[1;32m    101\u001b[0m values \u001b[38;5;241m=\u001b[39m [\n\u001b[1;32m    102\u001b[0m     (write\u001b[38;5;241m.\u001b[39mchannel, val)\n\u001b[1;32m    103\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m val, write \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mzip\u001b[39m(values, entries)\n\u001b[1;32m    104\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m write\u001b[38;5;241m.\u001b[39mskip_none \u001b[38;5;129;01mor\u001b[39;00m val \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[1;32m    105\u001b[0m ]\n\u001b[1;32m    106\u001b[0m \u001b[38;5;66;03m# write packets and values\u001b[39;00m\n",
+      "File \u001b[0;32m~/miniconda/lib/python3.9/site-packages/langgraph/pregel/write.py:98\u001b[0m, in \u001b[0;36m<listcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m     93\u001b[0m \u001b[38;5;66;03m# process entries into values\u001b[39;00m\n\u001b[1;32m     94\u001b[0m values \u001b[38;5;241m=\u001b[39m [\n\u001b[1;32m     95\u001b[0m     \u001b[38;5;28minput\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m write\u001b[38;5;241m.\u001b[39mvalue \u001b[38;5;129;01mis\u001b[39;00m PASSTHROUGH \u001b[38;5;28;01melse\u001b[39;00m write\u001b[38;5;241m.\u001b[39mvalue \u001b[38;5;28;01mfor\u001b[39;00m write \u001b[38;5;129;01min\u001b[39;00m entries\n\u001b[1;32m     96\u001b[0m ]\n\u001b[1;32m     97\u001b[0m values \u001b[38;5;241m=\u001b[39m [\n\u001b[0;32m---> 98\u001b[0m     val \u001b[38;5;28;01mif\u001b[39;00m write\u001b[38;5;241m.\u001b[39mmapper \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;28;01melse\u001b[39;00m \u001b[43mwrite\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mmapper\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43minvoke\u001b[49m\u001b[43m(\u001b[49m\u001b[43mval\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mconfig\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     99\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m val, write \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mzip\u001b[39m(values, entries)\n\u001b[1;32m    100\u001b[0m ]\n\u001b[1;32m    101\u001b[0m values \u001b[38;5;241m=\u001b[39m [\n\u001b[1;32m    102\u001b[0m     (write\u001b[38;5;241m.\u001b[39mchannel, val)\n\u001b[1;32m    103\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m val, write \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mzip\u001b[39m(values, entries)\n\u001b[1;32m    104\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m write\u001b[38;5;241m.\u001b[39mskip_none \u001b[38;5;129;01mor\u001b[39;00m val \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[1;32m    105\u001b[0m ]\n\u001b[1;32m    106\u001b[0m \u001b[38;5;66;03m# write packets and values\u001b[39;00m\n",
+      "File \u001b[0;32m~/miniconda/lib/python3.9/site-packages/langgraph/utils.py:95\u001b[0m, in \u001b[0;36mRunnableCallable.invoke\u001b[0;34m(self, input, config, **kwargs)\u001b[0m\n\u001b[1;32m     93\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m accepts_config(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mfunc):\n\u001b[1;32m     94\u001b[0m         kwargs[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mconfig\u001b[39m\u001b[38;5;124m\"\u001b[39m] \u001b[38;5;241m=\u001b[39m config\n\u001b[0;32m---> 95\u001b[0m     ret \u001b[38;5;241m=\u001b[39m \u001b[43mcontext\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mrun\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfunc\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43minput\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     96\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(ret, Runnable) \u001b[38;5;129;01mand\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mrecurse:\n\u001b[1;32m     97\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m ret\u001b[38;5;241m.\u001b[39minvoke(\u001b[38;5;28minput\u001b[39m, config)\n",
+      "File \u001b[0;32m~/miniconda/lib/python3.9/site-packages/langgraph/graph/state.py:300\u001b[0m, in \u001b[0;36mCompiledStateGraph.attach_node.<locals>._get_state_key\u001b[0;34m(input, config, key)\u001b[0m\n\u001b[1;32m    298\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m value \u001b[38;5;28;01mif\u001b[39;00m value \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;28;01melse\u001b[39;00m SKIP_WRITE\n\u001b[1;32m    299\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m--> 300\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m InvalidUpdateError(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mExpected dict, got \u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28minput\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m)\n",
+      "\u001b[0;31mInvalidUpdateError\u001b[0m: Expected dict, got Multiplier (since they provided the result of the multiplication)"
+     ]
+    }
+   ],
+   "source": [
+    "for s in graph.stream(\n",
+    "    {\n",
+    "        \"messages\": [\n",
+    "            HumanMessage(content=\"What is 4 multiplied by 35\")\n",
+    "        ]\n",
+    "    }\n",
+    "):\n",
+    "    if \"__end__\" not in s:\n",
+    "        print(s)\n",
+    "        print(\"----\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "6ff7b4d7-189a-4c31-9366-c42e3a0a7d47",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "2d14d9f3-bcc8-476a-b5f9-34987183636c",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.5"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

perm/CustomLLMMistral.py

@@ -0,0 +1,58 @@
+from langchain.llms.base import LLM
+from langchain.callbacks.manager import CallbackManagerForLLMRun
+from typing import Optional, List, Mapping, Any
+
+import warnings
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
+from transformers.models.mistral.modeling_mistral import MistralForCausalLM
+from transformers.models.llama.tokenization_llama_fast import LlamaTokenizerFast
+from pydantic import Field
+
+class CustomLLMMistral(LLM):
+    model: MistralForCausalLM = Field(...)
+    tokenizer: LlamaTokenizerFast = Field(...)
+    
+    def __init__(self):
+        
+        model_name = "mistralai/Mistral-7B-Instruct-v0.3"
+
+        quantization_config = BitsAndBytesConfig(load_in_4bit=True)
+        model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.bfloat16, quantization_config=quantization_config, device_map="auto")
+        tokenizer = AutoTokenizer.from_pretrained(model_name)
+        super().__init__(model=model, tokenizer=tokenizer)
+        self.model = model
+        self.tokenizer = tokenizer
+        
+
+    @property
+    def _llm_type(self) -> str:
+        return "custom"
+
+    def _call(self, prompt: str, stop: Optional[List[str]] = None,
+        run_manager: Optional[CallbackManagerForLLMRun] = None) -> str:
+
+        messages = [
+         {"role": "user", "content": prompt},
+        ]
+
+        encodeds = self.tokenizer.apply_chat_template(messages, return_tensors="pt")
+        model_inputs = encodeds.to(self.model.device)
+
+        generated_ids = self.model.generate(model_inputs, max_new_tokens=512, do_sample=True, pad_token_id=self.tokenizer.eos_token_id, top_k=4, temperature=0.7)
+        decoded = self.tokenizer.batch_decode(generated_ids)
+
+        output = decoded[0].split("[/INST]")[1].replace("</s>", "").strip()
+
+        if stop is not None:
+          for word in stop:
+            output = output.split(word)[0].strip()
+
+        while not output.endswith("```"):
+          output += "`"
+
+        return output
+
+    @property
+    def _identifying_params(self) -> Mapping[str, Any]:
+        return {"model": self.model}

perm/HuggingFaceAI.py

@@ -0,0 +1,22 @@
+from langchain_huggingface import HuggingFacePipeline, ChatHuggingFace
+from langchain_core.messages import BaseMessage
+from typing import List
+
+class HuggingFaceAI(ChatHuggingFace):
+    
+    def _to_chat_prompt(
+        self,
+        messages: List[BaseMessage],
+    ) -> str:
+        """Convert a list of messages into a prompt format expected by wrapped LLM."""
+        if not messages:
+            raise ValueError("At least one HumanMessage must be provided!")
+
+        if not isinstance(messages[-1], HumanMessage) and not isinstance(messages[-1], SystemMessage) :
+            raise ValueError("Last message must be a HumanMessage or SystemMessage!!!")
+
+        messages_dicts = [self._to_chatml_format(m) for m in messages]
+
+        return self.tokenizer.apply_chat_template(
+            messages_dicts, tokenize=False, add_generation_prompt=True
+        )

perm/agent_system.py

@@ -0,0 +1,55 @@
+from langgraph.graph import StateGraph, END
+from agents.agent_state import AgentState
+
+
+class AgentSystem:
+    def __init__(self, nodes, members):
+        self.workflow = self.setup_nodes(nodes)
+        self.setup_graph(members)
+        self.members = members
+   
+
+    def get_workflow(self, ):
+        return workflow
+    
+    def compile(self):
+        return self.workflow.compile()
+
+    def setup_nodes(self, nodes):
+        workflow = StateGraph(AgentState)
+
+        for node in nodes:
+            workflow.add_node(node["name"], node["instance"])
+        
+
+        return workflow
+
+    def check_step(self, x):
+        o = x["next"]
+        #print(f"Check for {o}")
+        #print(self.members)
+        
+        for member in self.members:
+            if member in o:
+                #print(f"Found member: {member} in target_string")
+                return member
+            #else:
+                #print(f"Member: {member} not found in target_string")
+                
+        #print("nothing found, Finish")
+        return "FINISH"
+        
+        
+    def setup_graph(self, members):
+        for member in members:
+         # We want our workers to ALWAYS "report back" to the supervisor when done
+            print(f"Add return path to supervisor for {member}")
+            self.workflow.add_edge(member, "supervisor")
+            
+        # The supervisor populates the "next" field in the graph state
+        # which routes to a node or finishes
+        conditional_map = {k: k for k in members}
+        conditional_map["FINISH"] = END
+        self.workflow.add_conditional_edges("supervisor", lambda x : self.check_step( x ), conditional_map)
+        # Finally, add entrypoint
+        self.workflow.set_entry_point("supervisor")

perm/agents/.ipynb_checkpoints/agent_node-checkpoint.py

@@ -0,0 +1,7 @@
+from langchain_core.messages import BaseMessage, HumanMessage
+
+def agent_node(state, agent, name):
+    result = agent.invoke(state)
+    return {"messages": [HumanMessage(content=result["output"], name=name)]}
+
+

perm/agents/.ipynb_checkpoints/agent_state-checkpoint.py

@@ -0,0 +1,11 @@
+from typing import Annotated, Any, Dict, List, Optional, Sequence, TypedDict
+from langchain_core.messages import BaseMessage
+import operator
+
+# The agent state is the input to each node in the graph
+class AgentState(TypedDict):
+    # The annotation tells the graph that new messages will always
+    # be added to the current states
+    messages: Annotated[Sequence[BaseMessage], operator.add]
+    # The 'next' field indicates where to route to next
+    next: str

perm/agents/.ipynb_checkpoints/agent_support-checkpoint.py

@@ -0,0 +1,23 @@
+from langchain.agents import AgentExecutor, create_openai_tools_agent
+from langchain_core.messages import BaseMessage, HumanMessage, SystemMessage, AIMessage
+from langchain_openai import ChatOpenAI
+
+from langchain_huggingface import ChatHuggingFace
+from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
+
+def create_agent(llm: ChatHuggingFace, tools: list, system_prompt: str):
+    # Each worker node will be given a name and some tools.
+    prompt = ChatPromptTemplate.from_messages(
+        [
+            (
+                "human",
+                system_prompt,
+            ),
+            ("assistant", "ok" ),
+            MessagesPlaceholder(variable_name="messages"),
+            MessagesPlaceholder(variable_name="agent_scratchpad"),
+        ]
+    )
+    agent = create_openai_tools_agent(llm, tools, prompt)
+    executor = AgentExecutor(agent=agent, tools=tools)
+    return executor

perm/agents/.ipynb_checkpoints/help_agent-checkpoint.py

@@ -0,0 +1,30 @@
+from langchain.agents import AgentExecutor, Agent, create_openai_tools_agent
+from langchain_core.prompts import PromptTemplate, ChatPromptTemplate, MessagesPlaceholder
+from langchain_core.tools import tool
+from langchain_openai import ChatOpenAI
+
+@tool
+def multiply(a: int, b: int) -> int:
+    """Multiply two numbers."""
+    print("Multiply used")
+    return a * b
+
+class HelpAgent(AgentExecutor):
+    
+    def __init__(self, llm, system_prompt):
+        prompt = ChatPromptTemplate.from_messages(
+            [
+                (
+                    "human",
+                    system_prompt,
+                ),
+                ("assistant",  "ok" ),  
+                MessagesPlaceholder(variable_name="messages"),
+                MessagesPlaceholder(variable_name="agent_scratchpad"),
+            ]
+        )
+        #agent = prompt | llm
+        agent = create_openai_tools_agent(llm, [multiply], prompt)
+        super().__init__(agent=agent, tools=[multiply])
+        
+    

perm/agents/.ipynb_checkpoints/project_agent-checkpoint.py

@@ -0,0 +1,41 @@
+from langchain.agents import AgentExecutor, Agent, create_openai_tools_agent
+from langchain_core.prompts import PromptTemplate, ChatPromptTemplate, MessagesPlaceholder
+from langchain_core.tools import tool
+from langchain_openai import ChatOpenAI
+#from tools.robot_information import robot_information
+from langchain.agents import create_tool_calling_agent
+from langchain.agents import AgentExecutor, ZeroShotAgent
+from langchain_core.tools import tool
+import json
+
+@tool
+def robot_information(project_name: str) -> str:
+    """Retrieves count and detailed information about the robots in the named project in real-time in JSON format"""
+    print("retrieved robot")
+    data = {
+        "count": 2,
+        "information": [
+            {"name": "Robot1", "battery": 90, "type": "heavy"},
+            {"name": "Robot2", "battery": 34, "type": "medium"}
+        ]
+    }
+    return json.dumps(data)
+class ProjectAgent(AgentExecutor):
+    
+    def __init__(self, llm, system_prompt):
+        prompt = ChatPromptTemplate.from_messages(
+            [
+               
+                MessagesPlaceholder(variable_name="messages"),
+                MessagesPlaceholder(variable_name="agent_scratchpad"),
+            ]
+        )
+        #agent = prompt | llm
+        tools = [robot_information] 
+        #llm_with_tools = llm.bind_tools(tools)
+       # agent = create_openai_tools_agent(llm, [robot_information], prompt)
+        agent = create_tool_calling_agent(llm, tools, prompt)
+       
+        #agent = prompt | llm_with_tools
+        super().__init__(agent=agent, tools=[robot_information], verbose=True)
+        

perm/agents/.ipynb_checkpoints/supervisor-checkpoint.py

@@ -0,0 +1,65 @@
+from langchain_core.output_parsers import StrOutputParser
+from langchain_core.prompts import PromptTemplate, ChatPromptTemplate, MessagesPlaceholder
+from langchain_core.runnables import RunnableLambda, RunnableSequence
+from langchain_core.messages import HumanMessage
+
+
+def strip_prompt(info):
+    print(info)
+    eot_token = "[/INST] "
+    i = info.content.rfind(eot_token)
+    if i == -1:
+        return info
+
+    info.content = info.content[i + len(eot_token):]
+
+    return {"next": info.content}
+   
+
+class Supervisor():
+
+    members = []
+    
+    system_prompt = (
+        "You are a supervisor tasked with managing a conversation between the"
+        " following workers:  {members}. Given the following user request,"
+        " respond with the worker to act next. Each worker will perform a"
+        " task and respond with their results and status. When finished,"
+        " respond with FINISH."
+    )
+
+    def __init__(self, llm, members):
+        self.members += members
+        self.prompt = ChatPromptTemplate.from_messages(
+            [
+                ("human", self.system_prompt),
+                ("assistant", "ok"),
+                MessagesPlaceholder(variable_name="messages"),
+                ("assistant", "ok"),
+                (
+                    "human",
+                    "Given the conversation above, who should act next?"
+                    " Or should we FINISH? Select one of: {options}",
+                ),
+            ]
+        ).partial(options=str(self.get_options()), members=", ".join(self.members))
+
+        self.chain = ( self.prompt | llm | RunnableLambda(strip_prompt))
+
+    def add_member(self, member):
+        self.members.append(member)
+
+    def get_members(self):
+        return self.members;
+        
+    def get_options(self):
+        return ["FINISH"] + self.members
+
+    def get_chain(self):
+        return self.chain
+        
+
+    def invoke(self, query):
+        self.chain.invoke([HumanMessage(query)])
+
+

perm/agents/agent_node.py

@@ -0,0 +1,7 @@
+from langchain_core.messages import BaseMessage, HumanMessage
+
+def agent_node(state, agent, name):
+    result = agent.invoke(state)
+    return {"messages": [HumanMessage(content=result["output"], name=name)]}
+
+

perm/agents/agent_state.py

@@ -0,0 +1,11 @@
+from typing import Annotated, Any, Dict, List, Optional, Sequence, TypedDict
+from langchain_core.messages import BaseMessage
+import operator
+
+# The agent state is the input to each node in the graph
+class AgentState(TypedDict):
+    # The annotation tells the graph that new messages will always
+    # be added to the current states
+    messages: Annotated[Sequence[BaseMessage], operator.add]
+    # The 'next' field indicates where to route to next
+    next: str

perm/agents/agent_support.py

@@ -0,0 +1,23 @@
+from langchain.agents import AgentExecutor, create_openai_tools_agent
+from langchain_core.messages import BaseMessage, HumanMessage, SystemMessage, AIMessage
+from langchain_openai import ChatOpenAI
+
+from langchain_huggingface import ChatHuggingFace
+from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
+
+def create_agent(llm: ChatHuggingFace, tools: list, system_prompt: str):
+    # Each worker node will be given a name and some tools.
+    prompt = ChatPromptTemplate.from_messages(
+        [
+            (
+                "human",
+                system_prompt,
+            ),
+            ("assistant", "ok" ),
+            MessagesPlaceholder(variable_name="messages"),
+            MessagesPlaceholder(variable_name="agent_scratchpad"),
+        ]
+    )
+    agent = create_openai_tools_agent(llm, tools, prompt)
+    executor = AgentExecutor(agent=agent, tools=tools)
+    return executor

perm/agents/help_agent.py

@@ -0,0 +1,30 @@
+from langchain.agents import AgentExecutor, Agent, create_openai_tools_agent
+from langchain_core.prompts import PromptTemplate, ChatPromptTemplate, MessagesPlaceholder
+from langchain_core.tools import tool
+from langchain_openai import ChatOpenAI
+
+@tool
+def multiply(a: int, b: int) -> int:
+    """Multiply two numbers."""
+    print("Multiply used")
+    return a * b
+
+class HelpAgent(AgentExecutor):
+    
+    def __init__(self, llm, system_prompt):
+        prompt = ChatPromptTemplate.from_messages(
+            [
+                (
+                    "human",
+                    system_prompt,
+                ),
+                ("assistant",  "ok" ),  
+                MessagesPlaceholder(variable_name="messages"),
+                MessagesPlaceholder(variable_name="agent_scratchpad"),
+            ]
+        )
+        #agent = prompt | llm
+        agent = create_openai_tools_agent(llm, [multiply], prompt)
+        super().__init__(agent=agent, tools=[multiply])
+        
+    

perm/agents/project_agent.py

@@ -0,0 +1,41 @@
+from langchain.agents import AgentExecutor, Agent, create_openai_tools_agent
+from langchain_core.prompts import PromptTemplate, ChatPromptTemplate, MessagesPlaceholder
+from langchain_core.tools import tool
+from langchain_openai import ChatOpenAI
+#from tools.robot_information import robot_information
+from langchain.agents import create_tool_calling_agent
+from langchain.agents import AgentExecutor, ZeroShotAgent
+from langchain_core.tools import tool
+import json
+
+@tool
+def robot_information(project_name: str) -> str:
+    """Retrieves count and detailed information about the robots in the named project in real-time in JSON format"""
+    print("retrieved robot")
+    data = {
+        "count": 2,
+        "information": [
+            {"name": "Robot1", "battery": 90, "type": "heavy"},
+            {"name": "Robot2", "battery": 34, "type": "medium"}
+        ]
+    }
+    return json.dumps(data)
+class ProjectAgent(AgentExecutor):
+    
+    def __init__(self, llm, system_prompt):
+        prompt = ChatPromptTemplate.from_messages(
+            [
+               
+                MessagesPlaceholder(variable_name="messages"),
+                MessagesPlaceholder(variable_name="agent_scratchpad"),
+            ]
+        )
+        #agent = prompt | llm
+        tools = [robot_information] 
+        #llm_with_tools = llm.bind_tools(tools)
+       # agent = create_openai_tools_agent(llm, [robot_information], prompt)
+        agent = create_tool_calling_agent(llm, tools, prompt)
+       
+        #agent = prompt | llm_with_tools
+        super().__init__(agent=agent, tools=[robot_information], verbose=True)
+        

perm/agents/supervisor.py

@@ -0,0 +1,65 @@
+from langchain_core.output_parsers import StrOutputParser
+from langchain_core.prompts import PromptTemplate, ChatPromptTemplate, MessagesPlaceholder
+from langchain_core.runnables import RunnableLambda, RunnableSequence
+from langchain_core.messages import HumanMessage
+
+
+def strip_prompt(info):
+    print(info)
+    eot_token = "[/INST] "
+    i = info.content.rfind(eot_token)
+    if i == -1:
+        return info
+
+    info.content = info.content[i + len(eot_token):]
+
+    return {"next": info.content}
+   
+
+class Supervisor():
+
+    members = []
+    
+    system_prompt = (
+        "You are a supervisor tasked with managing a conversation between the"
+        " following workers:  {members}. Given the following user request,"
+        " respond with the worker to act next. Each worker will perform a"
+        " task and respond with their results and status. When finished,"
+        " respond with FINISH."
+    )
+
+    def __init__(self, llm, members):
+        self.members += members
+        self.prompt = ChatPromptTemplate.from_messages(
+            [
+                ("human", self.system_prompt),
+                ("assistant", "ok"),
+                MessagesPlaceholder(variable_name="messages"),
+                ("assistant", "ok"),
+                (
+                    "human",
+                    "Given the conversation above, who should act next?"
+                    " Or should we FINISH? Select one of: {options}",
+                ),
+            ]
+        ).partial(options=str(self.get_options()), members=", ".join(self.members))
+
+        self.chain = ( self.prompt | llm | RunnableLambda(strip_prompt))
+
+    def add_member(self, member):
+        self.members.append(member)
+
+    def get_members(self):
+        return self.members;
+        
+    def get_options(self):
+        return ["FINISH"] + self.members
+
+    def get_chain(self):
+        return self.chain
+        
+
+    def invoke(self, query):
+        self.chain.invoke([HumanMessage(query)])
+
+

perm/app.py

@@ -0,0 +1,58 @@
+import HuggingFaceAI
+from langchain_huggingface import  HuggingFacePipeline, ChatHuggingFace
+from langchain_core.messages import BaseMessage, HumanMessage, AIMessage
+
+from agents.supervisor import Supervisor
+from agents.agent_support import create_agent
+from agents.agent_node import agent_node
+
+from agents.help_agent import HelpAgent
+from agents.project_agent import ProjectAgent
+
+from agent_system import AgentSystem
+
+from tools.multiply_tool import multiply
+import functools
+
+import os
+from uuid import uuid4
+
+unique_id = uuid4().hex[0:8]
+os.environ["LANGCHAIN_TRACING_V2"] = "true"
+os.environ["LANGCHAIN_PROJECT"] = f"InfiniFleetTrace"
+os.environ["LANGCHAIN_ENDPOINT"] = "https://api.smith.langchain.com"
+os.environ["LANGCHAIN_API_KEY"] = "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"  # Update to your API key
+
+
+llm = HuggingFacePipeline.from_model_id(
+    model_id="mistralai/Mistral-7B-Instruct-v0.3",
+    device_map="auto",
+    task="text-generation",
+    pipeline_kwargs={
+        "max_new_tokens": 100,
+        "top_k": 50,
+    },
+)
+
+print("Creating chat interface")
+chat_llm = ChatHuggingFace(llm=llm)
+print("Done")
+print("---------")
+
+
+supervisor = Supervisor(chat_llm, ["ProductHelp","ProjectHelp", "Multiplier"])
+
+
+help_agent = HelpAgent(chat_llm, "You provide help for the InfiniFleet product in general")
+help_node = functools.partial(agent_node, agent=help_agent, name="ProductHelp")
+
+project_agent = ProjectAgent(chat_llm, "Always use robot_information tool to get all required information.")
+project_node = functools.partial(agent_node, agent=project_agent, name="ProjectHelp")
+
+print("--project agent-------")
+input_data = "Use the tool to give me information about how many robots there are in the project called 'largeProject'."
+result = project_agent.invoke({"messages":[HumanMessage(input_data)]})
+print(result)
+print("---------")
+
+

perm/experiment.py

@@ -0,0 +1,77 @@
+from CustomLLMMistral import CustomLLMMistral
+from tools.robot_information import robot_information
+import os
+
+os.environ["LANGCHAIN_TRACING_V2"] = "true"
+os.environ["LANGCHAIN_PROJECT"] = f"InfiniFleetTrace"
+os.environ["LANGCHAIN_ENDPOINT"] = "https://api.smith.langchain.com"
+os.environ["LANGCHAIN_API_KEY"] = "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"  # Update to your API key
+
+llm = CustomLLMMistral()
+
+#info = robot_information.invoke("test")
+#print(info)
+
+tools = [ robot_information ]
+
+system="""
+You are designed to solve tasks. Each task requires multiple steps that are represented by a markdown code snippet of a json blob.
+The json structure should contain the following keys:
+thought -> your thoughts
+action -> name of a tool
+action_input -> parameters to send to the tool
+
+These are the tools you can use: {tool_names}.
+
+These are the tools descriptions:
+
+{tools}
+
+If you have enough information to answer the query use the tool "Final Answer". Its parameters is the solution.
+If there is not enough information, keep trying.
+
+"""
+
+human="""
+Add the word "STOP" after each markdown snippet. Example:
+
+```json
+{{"thought": "<your thoughts>",
+ "action": "<tool name or Final Answer to give a final answer>",
+ "action_input": "<tool parameters or the final output"}}
+```
+STOP
+
+This is my query="{input}". Write only the next step needed to solve it.
+Your answer should be based in the previous tools executions, even if you think you know the answer.
+Remember to add STOP after each snippet.
+
+These were the previous steps given to solve this query and the information you already gathered:
+"""
+
+from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
+
+prompt = ChatPromptTemplate.from_messages(
+    [
+        ("system", system),
+        MessagesPlaceholder("chat_history", optional=True),
+        ("human", human),
+        MessagesPlaceholder("agent_scratchpad"),
+    ]
+)
+
+from langchain.agents import create_json_chat_agent, AgentExecutor
+from langchain.memory import ConversationBufferMemory
+
+agent = create_json_chat_agent(
+    tools = tools,
+    llm = llm,
+    prompt = prompt,
+    stop_sequence = ["STOP"],
+    template_tool_response = "{observation}"
+)
+
+memory = AgentExecutor(agent=agent, tools=tools, verbose=True)
+agent_executor = AgentExecutor(agent=agent, tools=tools, verbose=True, handle_parsing_errors=True)
+
+agent_executor.invoke({"input": "Who are you?"})

perm/hello.py

@@ -0,0 +1,3 @@
+from langchain_huggingface import HuggingFacePipeline
+print("Hello world")
+

perm/scratchpad.py

@@ -0,0 +1,25 @@
+multiply_agent = create_agent(chat_llm, [multiply], "You multiply numbers")
+multiply_node = functools.partial(agent_node, agent=multiply_agent, name="Multiplier")
+
+
+nodes = [
+    { "name": "supervisor", "instance": supervisor.get_chain() },
+    { "name": "ProductHelp", "instance": help_node },
+    { "name": "ProjectHelp", "instance": project_node },
+    { "name": "Multiplier", "instance": multiply_node }
+]
+
+
+agent_system = AgentSystem(nodes, supervisor.get_members())
+graph = agent_system.compile()
+
+for s in graph.stream(
+    {
+        "messages": [
+            HumanMessage(content="How many robots are currently set up in the project?")
+        ]
+    }
+):
+    if "__end__" not in s:
+        print(s)
+        print("----")

perm/start.sh

@@ -0,0 +1,5 @@
+export env LANGCHAIN_TRACING_V2=true
+export env LANGCHAIN_ENDPOINT="https://api.smith.langchain.com"
+export env LANGCHAIN_API_KEY=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
+export env LANGCHAIN_PROJECT=TestProject
+pip install -U "huggingface_hub[cli]"

perm/tasks.json

@@ -0,0 +1,10 @@
+{
+	"version": "2.0.0",
+	"tasks": [
+		{
+			"label": "Env & Login",
+			"type": "shell",
+			"command": "bash ./start.sh"
+		}
+	]
+}

perm/tools/.ipynb_checkpoints/multiply_tool-checkpoint.py

@@ -0,0 +1,7 @@
+from langchain_core.tools import tool
+
+@tool
+def multiply(a: int, b: int) -> int:
+    """Multiply two numbers."""
+    print("Multiply used")
+    return a * b

perm/tools/.ipynb_checkpoints/robot_information-checkpoint.py

@@ -0,0 +1,14 @@
+from langchain_core.tools import tool
+import json
+
+@tool
+def robot_information(project_name: str) -> str:
+    """Retrieves detailed information about the robots in the named project in real-time in JSON format"""
+    print("retrieved robot")
+    data = {
+        "information": [
+            {"name": "Robot1", "battery": 90, "type": "heavy"},
+            {"name": "Robot2", "battery": 34, "type": "medium"}
+        ]
+    }
+    return json.dumps(data)

perm/tools/multiply_tool.py

@@ -0,0 +1,7 @@
+from langchain_core.tools import tool
+
+@tool
+def multiply(a: int, b: int) -> int:
+    """Multiply two numbers."""
+    print("Multiply used")
+    return a * b

perm/tools/robot_information.py

@@ -0,0 +1,14 @@
+from langchain_core.tools import tool
+import json
+
+@tool
+def robot_information(project_name: str) -> str:
+    """Retrieves detailed information about the robots in the named project in real-time in JSON format"""
+    print("retrieved robot")
+    data = {
+        "information": [
+            {"name": "Robot1", "battery": 90, "type": "heavy"},
+            {"name": "Robot2", "battery": 34, "type": "medium"}
+        ]
+    }
+    return json.dumps(data)