Added sample_generator.ipynb.

demo/sample_generator.ipynb

@@ -0,0 +1,196 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import os\n",
+    "\n",
+    "# Load configuration from YAML file\n",
+    "config = {\n",
+    "    \"model_name\": \"llama3-70b-8192\",\n",
+    "}\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/yale/work/meta-prompt/.venv/lib/python3.10/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
+      "  from .autonotebook import tqdm as notebook_tqdm\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Running on local URL:  http://127.0.0.1:7861\n",
+      "\n",
+      "To create a public link, set `share=True` in `launch()`.\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<div><iframe src=\"http://127.0.0.1:7861/\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
+      ],
+      "text/plain": [
+       "<IPython.core.display.HTML object>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/yale/work/meta-prompt/.venv/lib/python3.10/site-packages/langchain_core/_api/deprecation.py:141: LangChainDeprecationWarning: The class `ChatOpenAI` was deprecated in LangChain 0.0.10 and will be removed in 0.3.0. An updated version of the class exists in the langchain-openai package and should be used instead. To use it run `pip install -U langchain-openai` and import as `from langchain_openai import ChatOpenAI`.\n",
+      "  warn_deprecated(\n"
+     ]
+    }
+   ],
+   "source": [
+    "import json\n",
+    "import yaml\n",
+    "from langchain.prompts import ChatPromptTemplate\n",
+    "from langchain.chat_models import ChatOpenAI\n",
+    "from langchain.schema.output_parser import StrOutputParser\n",
+    "\n",
+    "class TaskDescriptionGenerator:\n",
+    "    def __init__(self, model_temperature=0.7):\n",
+    "        self.model = ChatOpenAI(model=config[\"model_name\"], temperature=model_temperature)\n",
+    "        self.output_parser = StrOutputParser()\n",
+    "        \n",
+    "        self.description_prompt = ChatPromptTemplate.from_template(\n",
+    "            \"\"\"Given the following JSON example for a task type:\n",
+    "            {raw_example}\n",
+    "\n",
+    "            Provide a concise description of the task type.\n",
+    "\n",
+    "            Format your response as follows:\n",
+    "            Task Description: [Your description here]\n",
+    "            \"\"\"\n",
+    "        )\n",
+    "\n",
+    "        self.examples_prompt = ChatPromptTemplate.from_template(\n",
+    "            \"\"\"Given the following task type description and JSON example:\n",
+    "            Task Description: {description}\n",
+    "            \n",
+    "            Example:\n",
+    "            {raw_example}\n",
+    "\n",
+    "            Generate 3 more input/output examples for this task type in the same JSON format.\n",
+    "\n",
+    "            Format your response as a valid JSON object with a single key 'examples' containing a JSON array of 3 objects, each with 'input' and 'output' fields.\n",
+    "            \"\"\"\n",
+    "        )\n",
+    "\n",
+    "        self.description_chain = self.description_prompt | self.model | self.output_parser\n",
+    "        # bind json_object to the model\n",
+    "        json_model = self.model.bind(response_format={\"type\": \"json_object\"})\n",
+    "        self.examples_chain = self.examples_prompt | json_model | self.output_parser\n",
+    "\n",
+    "    def generate_description(self, raw_example_str):\n",
+    "        result = self.description_chain.invoke({\n",
+    "            \"raw_example\": raw_example_str\n",
+    "        })\n",
+    "        return result.split(\"Task Description: \")[1].strip()\n",
+    "\n",
+    "    def generate_examples(self, description, raw_example_str):\n",
+    "        result = self.examples_chain.invoke({\n",
+    "            \"description\": description,\n",
+    "            \"raw_example\": raw_example_str\n",
+    "        })\n",
+    "\n",
+    "        try:\n",
+    "            result = result.strip()\n",
+    "            if result.startswith('```') and result.endswith('```'):\n",
+    "                result = result.strip('```').strip()\n",
+    "                if result.startswith('json'):\n",
+    "                    result = result.strip('json').strip()\n",
+    "            return json.loads(result)\n",
+    "        except json.JSONDecodeError as e:\n",
+    "            raise ValueError(f\"The generated examples are not in valid JSON format. Error: {str(e)} Result: {result}\")\n",
+    "\n",
+    "    def process(self, input_str):\n",
+    "        try:\n",
+    "            # Try to parse as JSON\n",
+    "            try:\n",
+    "                data = json.loads(input_str)\n",
+    "            except json.JSONDecodeError:\n",
+    "                # If JSON parsing fails, try to parse as YAML\n",
+    "                try:\n",
+    "                    data = yaml.safe_load(input_str)\n",
+    "                except yaml.YAMLError as e:\n",
+    "                    raise ValueError(\"Invalid input string. Error: {}\".format(str(e)))\n",
+    "            \n",
+    "            if not isinstance(data, dict) or 'input' not in data or 'output' not in data:\n",
+    "                raise ValueError(\"Invalid input format. Expected an object with 'input' and 'output' fields.\")\n",
+    "            \n",
+    "            description = self.generate_description(json.dumps(data))\n",
+    "            new_examples = self.generate_examples(description, json.dumps(data))\n",
+    "            \n",
+    "            output = {\n",
+    "                \"task_description\": description,\n",
+    "                \"additional_examples\": new_examples\n",
+    "            }\n",
+    "            \n",
+    "            return json.dumps(output, indent=2)\n",
+    "        \n",
+    "        except Exception as e:\n",
+    "            raise RuntimeError(f\"An error occurred during processing: {str(e)}\")\n",
+    "    \n",
+    "\n",
+    "import gradio as gr\n",
+    "\n",
+    "def process_json(input_json):\n",
+    "    try:\n",
+    "        generator = TaskDescriptionGenerator()\n",
+    "        result = generator.process(input_json)\n",
+    "        return \"Process completed successfully. Result:\\n\" + result\n",
+    "    except Exception as e:\n",
+    "        return f\"An error occurred: {str(e)}\"\n",
+    "\n",
+    "demo = gr.Interface(\n",
+    "    fn=process_json,\n",
+    "    inputs=gr.Textbox(label=\"Input JSON\"),\n",
+    "    outputs=gr.Textbox(label=\"Output\"),\n",
+    "    title=\"Task Description Generator\",\n",
+    "    description=\"Enter a JSON object with 'input' and 'output' fields to generate a task description and additional examples.\"\n",
+    ")\n",
+    "\n",
+    "if __name__ == \"__main__\":\n",
+    "    demo.launch()"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": ".venv",
+   "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": "3.10.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}