diff --git a/notebooks/adalflow_colab_template.ipynb b/notebooks/adalflow_colab_template.ipynb
index 39715816..746f12d3 100644
--- a/notebooks/adalflow_colab_template.ipynb
+++ b/notebooks/adalflow_colab_template.ipynb
@@ -58,6 +58,17 @@
"clear_output()"
]
},
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "!pip uninstall httpx anyio -y\n",
+ "!pip install \"anyio>=3.1.0,<4.0\"\n",
+ "!pip install httpx==0.24.1"
+ ]
+ },
{
"cell_type": "markdown",
"metadata": {},
diff --git a/notebooks/qas/adalflow_object_count_auto_optimization.ipynb b/notebooks/qas/adalflow_object_count_auto_optimization.ipynb
index 9308ea7f..533ba124 100644
--- a/notebooks/qas/adalflow_object_count_auto_optimization.ipynb
+++ b/notebooks/qas/adalflow_object_count_auto_optimization.ipynb
@@ -62,6 +62,17 @@
"clear_output()"
]
},
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "!pip uninstall httpx anyio -y\n",
+ "!pip install \"anyio>=3.1.0,<4.0\"\n",
+ "!pip install httpx==0.24.1"
+ ]
+ },
{
"cell_type": "code",
"execution_count": null,
diff --git a/notebooks/tutorials/adalflow_classification_optimization.ipynb b/notebooks/tutorials/adalflow_classification_optimization.ipynb
index c6bddc7e..36332387 100644
--- a/notebooks/tutorials/adalflow_classification_optimization.ipynb
+++ b/notebooks/tutorials/adalflow_classification_optimization.ipynb
@@ -1,461 +1,967 @@
{
- "nbformat": 4,
- "nbformat_minor": 0,
- "metadata": {
- "colab": {
- "provenance": []
- },
- "kernelspec": {
- "name": "python3",
- "display_name": "Python 3"
- },
- "language_info": {
- "name": "python"
- }
- },
- "cells": [
- {
- "cell_type": "markdown",
- "source": [
- "# ๐ค Welcome to AdalFlow!\n",
- "## The PyTorch library to auto-optimize any LLM task pipelines\n",
- "\n",
- "Thanks for trying us out, we're here to provide you with the best LLM application development experience you can dream of ๐ any questions or concerns you may have, [come talk to us on discord,](https://discord.gg/ezzszrRZvT) we're always here to help! โญ Star us on Github โญ\n",
- "\n",
- "\n",
- "# Quick Links\n",
- "\n",
- "Github repo: https://github.com/SylphAI-Inc/AdalFlow\n",
- "\n",
- "Full Tutorials: https://adalflow.sylph.ai/index.html#.\n",
- "\n",
- "Deep dive on each API: check out the [developer notes](https://adalflow.sylph.ai/tutorials/index.html).\n",
- "\n",
- "Common use cases along with the auto-optimization: check out [Use cases](https://adalflow.sylph.ai/use_cases/index.html).\n",
- "\n",
- "## ๐ Outline\n",
- "\n",
- "This is the code for a classification optimization tutorial \n"
- ],
- "metadata": {
- "id": "xHF95Kr4CzGq"
- }
- },
- {
- "cell_type": "markdown",
- "source": [
- "\n",
- "# Installation\n",
- "\n",
- "1. Use `pip` to install the `adalflow` Python package. We will need `openai`, `groq` from the extra packages.\n",
- "\n",
- " ```bash\n",
- " pip install adalflow[openai,groq]\n",
- " ```\n",
- "2. Setup `openai` and `groq` API key in the environment variables\n",
- "\n",
- "You can choose to use different client. You can import the model client you prefer. We support `Anthropic`, `Cohere`, `Google`, `GROQ`, `OpenAI`, `Transformer` and more in development. We will use OpenAI here as an example.Please refer to our [full installation guide](https://adalflow.sylph.ai/get_started/installation.html)"
- ],
- "metadata": {
- "id": "Kof5M6DRaKhh"
- }
- },
- {
- "cell_type": "code",
- "execution_count": 42,
- "metadata": {
- "id": "tAp3eDjOCma1"
- },
- "outputs": [],
- "source": [
- "from IPython.display import clear_output\n",
- "\n",
- "!pip install -U adalflow[openai] # also install the package for the model client you'll use\n",
- "!pip install datasets\n",
- "clear_output()"
- ]
- },
- {
- "cell_type": "markdown",
- "source": [
- "## Set Environment Variables\n",
- "\n",
- "Run the following code and pass your api key.\n",
- "\n",
- "Note: for normal `.py` projects, follow our [official installation guide](https://lightrag.sylph.ai/get_started/installation.html).\n",
- "\n",
- "*Go to [OpenAI](https://platform.openai.com/docs/introduction) to get API keys if you don't already have.*"
- ],
- "metadata": {
- "id": "KapUyHMM07pJ"
- }
- },
- {
- "cell_type": "code",
- "source": [
- "import os\n",
- "\n",
- "from getpass import getpass\n",
- "\n",
- "# Prompt user to enter their API keys securely\n",
- "openai_api_key = getpass(\"Please enter your OpenAI API key: \")\n",
- "\n",
- "\n",
- "# Set environment variables\n",
- "os.environ[\"OPENAI_API_KEY\"] = openai_api_key\n",
- "\n",
- "print(\"API keys have been set.\")"
- ],
- "metadata": {
+ "nbformat": 4,
+ "nbformat_minor": 0,
+ "metadata": {
"colab": {
- "base_uri": "https://localhost:8080/"
+ "provenance": []
},
- "id": "ONfzF9Puzdd_",
- "outputId": "e5c3cfc5-69cb-448a-c248-a8cebda5ba71"
- },
- "execution_count": 43,
- "outputs": [
- {
- "output_type": "stream",
- "name": "stdout",
- "text": [
- "Please enter your OpenAI API key: ยทยทยทยทยทยทยทยทยทยท\n",
- "API keys have been set.\n"
- ]
+ "kernelspec": {
+ "name": "python3",
+ "display_name": "Python 3"
+ },
+ "language_info": {
+ "name": "python"
}
- ]
- },
- {
- "cell_type": "code",
- "source": [
- "from dataclasses import dataclass, field\n",
- "from typing import List, Dict, Union, Optional, Tuple, Any, Callable\n",
- "from datasets import load_dataset\n",
- "from adalflow.components.model_client import OpenAIClient\n",
- "import adalflow as adal\n",
- "from adalflow.core.component import Component\n",
- "from adalflow.datasets.types import TrecData\n",
- "from adalflow.eval.answer_match_acc import AnswerMatchAcc\n",
- "\n",
- "\n",
- "_COARSE_LABELS = [\"ABBR\", \"DESC\", \"ENTY\", \"HUM\", \"LOC\", \"NUM\"]\n",
- "\n",
- "_COARSE_LABELS_DESC = [\n",
- " \"Abbreviation: Questions about abbreviations and their meanings\",\n",
- " \"Description: Questions seeking descriptions of people, things, or concepts\",\n",
- " \"Entity: Questions about entities (e.g., animals, colors, inventions)\",\n",
- " \"Human: Questions about people or organizations\",\n",
- " \"Location: Questions about places, cities, countries\",\n",
- " \"Numeric: Questions seeking numeric answers (e.g., dates, amounts, distances)\",\n",
- "]\n",
- "\n",
- "\n",
- "template = r\"\"\"\n",
- " {{system_prompt}}\n",
- " {% if output_format_str is not none %}\n",
- " {{output_format_str}}\n",
- " {% endif %}\n",
- " {% if few_shot_demos is not none %}\n",
- " Here are some examples:\n",
- " {{few_shot_demos}}\n",
- " {% endif %}\n",
- " \n",
- " \n",
- " {{input_str}}\n",
- " \n",
- " \"\"\"\n",
- "\n",
- "task_desc_template = r\"\"\"You are a classifier. Given a question, you need to classify it into one of the following classes:\n",
- " Format: class_index. class_name, class_description\n",
- " {% if classes %}\n",
- " {% for class in classes %}\n",
- " {{loop.index-1}}. {{class.label}}, {{class.desc}}\n",
- " {% endfor %}\n",
- " {% endif %}\n",
- " - Do not try to answer the question:\n",
- " \"\"\"\n",
- "\n",
- "\n",
- "@dataclass\n",
- "class TRECExtendedData(TrecData):\n",
- " rationale: str = field(\n",
- " metadata={\n",
- " \"desc\": \"Your step-by-step reasoning to classify the question to class_name\"\n",
- " },\n",
- " default=None,\n",
- " )\n",
- " __input_fields__ = [\"question\"]\n",
- " __output_fields__ = [\n",
- " \"rationale\",\n",
- " \"class_name\",\n",
- " ] # it is important to have the rationale before the class_name"
- ],
- "metadata": {
- "id": "ZZIEtZYHNVjo"
- },
- "execution_count": 49,
- "outputs": []
- },
- {
- "cell_type": "code",
- "source": [
- "class TRECClassifierStructuredOutput(adal.Component):\n",
- "\n",
- " def __init__(self, model_client: adal.ModelClient, model_kwargs: Dict):\n",
- " super().__init__()\n",
- "\n",
- " label_desc = [\n",
- " {\"label\": label, \"desc\": desc}\n",
- " for label, desc in zip(_COARSE_LABELS, _COARSE_LABELS_DESC)\n",
- " ]\n",
- "\n",
- " task_desc_str = adal.Prompt(\n",
- " template=task_desc_template, prompt_kwargs={\"classes\": label_desc}\n",
- " )()\n",
- "\n",
- " self.data_class = TRECExtendedData\n",
- " self.data_class.set_task_desc(task_desc_str)\n",
- "\n",
- " self.parser = adal.DataClassParser(\n",
- " data_class=self.data_class, return_data_class=True, format_type=\"yaml\"\n",
- " )\n",
- "\n",
- " prompt_kwargs = {\n",
- " \"system_prompt\": adal.Parameter(\n",
- " data=self.parser.get_task_desc_str(),\n",
- " role_desc=\"Task description\",\n",
- " requires_opt=True,\n",
- " param_type=adal.ParameterType.PROMPT,\n",
- " ),\n",
- " \"output_format_str\": adal.Parameter(\n",
- " data=self.parser.get_output_format_str(),\n",
- " role_desc=\"Output format requirements\",\n",
- " requires_opt=False,\n",
- " param_type=adal.ParameterType.PROMPT,\n",
- " ),\n",
- " \"few_shot_demos\": adal.Parameter(\n",
- " data=None,\n",
- " requires_opt=True,\n",
- " role_desc=\"Few shot examples to help the model\",\n",
- " param_type=adal.ParameterType.DEMOS,\n",
- " ),\n",
- " }\n",
- "\n",
- " self.llm = adal.Generator(\n",
- " model_client=model_client,\n",
- " model_kwargs=model_kwargs,\n",
- " prompt_kwargs=prompt_kwargs,\n",
- " template=template,\n",
- " output_processors=self.parser,\n",
- " use_cache=True,\n",
- " )\n",
- "\n",
- " def _prepare_input(self, question: str):\n",
- " input_data = self.data_class(question=question)\n",
- " input_str = self.parser.get_input_str(input_data)\n",
- " prompt_kwargs = {\n",
- " \"input_str\": adal.Parameter(\n",
- " data=input_str, requires_opt=False, role_desc=\"input to the LLM\"\n",
- " )\n",
- " }\n",
- " return prompt_kwargs\n",
- "\n",
- " def call(\n",
- " self, question: str, id: Optional[str] = None\n",
- " ) -> Union[adal.GeneratorOutput, adal.Parameter]:\n",
- " prompt_kwargs = self._prepare_input(question)\n",
- " output = self.llm(prompt_kwargs=prompt_kwargs, id=id)\n",
- " return output"
- ],
- "metadata": {
- "id": "3Q3H9XC4Ncfi"
- },
- "execution_count": 50,
- "outputs": []
- },
- {
- "cell_type": "code",
- "source": [
- "class TrecClassifierAdal(adal.AdalComponent):\n",
- " def __init__(\n",
- " self,\n",
- " model_client: adal.ModelClient,\n",
- " model_kwargs: Dict,\n",
- " teacher_model_config: Dict,\n",
- " backward_engine_model_config: Dict,\n",
- " text_optimizer_model_config: Dict,\n",
- " ):\n",
- " task = TRECClassifierStructuredOutput(model_client, model_kwargs)\n",
- " eval_fn = AnswerMatchAcc(type=\"exact_match\").compute_single_item\n",
- " loss_fn = adal.EvalFnToTextLoss(\n",
- " eval_fn=eval_fn,\n",
- " eval_fn_desc=\"exact_match: 1 if str(y) == str(y_gt) else 0\",\n",
- " )\n",
- " super().__init__(\n",
- " task=task,\n",
- " eval_fn=eval_fn,\n",
- " loss_fn=loss_fn,\n",
- " backward_engine_model_config=backward_engine_model_config,\n",
- " text_optimizer_model_config=text_optimizer_model_config,\n",
- " teacher_model_config=teacher_model_config,\n",
- " )\n",
- "\n",
- " def prepare_task(self, sample: TRECExtendedData):\n",
- " return self.task.call, {\"question\": sample.question, \"id\": sample.id}\n",
- "\n",
- " def prepare_eval(\n",
- " self, sample: TRECExtendedData, y_pred: adal.GeneratorOutput\n",
- " ) -> float:\n",
- " y_label = -1\n",
- " if y_pred and y_pred.data is not None and y_pred.data.class_name is not None:\n",
- " y_label = y_pred.data.class_name\n",
- " return self.eval_fn, {\"y\": y_label, \"y_gt\": sample.class_name}\n",
- "\n",
- " def prepare_loss(\n",
- " self, sample: TRECExtendedData, y_pred: adal.Parameter, *args, **kwargs\n",
- " ) -> Tuple[Callable[..., Any], Dict]:\n",
- " full_response = y_pred.full_response\n",
- " y_label = -1\n",
- " if (\n",
- " full_response\n",
- " and full_response.data is not None\n",
- " and full_response.data.class_name is not None\n",
- " ):\n",
- " y_label = full_response.data.class_name\n",
- "\n",
- " y_pred.eval_input = y_label\n",
- " y_gt = adal.Parameter(\n",
- " name=\"y_gt\",\n",
- " data=sample.class_name,\n",
- " eval_input=sample.class_name,\n",
- " requires_opt=False,\n",
- " )\n",
- " return self.loss_fn, {\"kwargs\": {\"y\": y_pred, \"y_gt\": y_gt}}"
- ],
- "metadata": {
- "id": "HpkQYsh2NevT"
- },
- "execution_count": 51,
- "outputs": []
- },
- {
- "cell_type": "code",
- "source": [
- "def train(\n",
- " model_client: adal.ModelClient,\n",
- " model_kwargs: Dict,\n",
- " train_batch_size=4,\n",
- " raw_shots: int = 0,\n",
- " bootstrap_shots: int = 1,\n",
- " max_steps=12,\n",
- " num_workers=4,\n",
- " strategy=\"constrained\",\n",
- " optimization_order=\"sequential\",\n",
- " debug=False,\n",
- "):\n",
- " print(\"Starting training process...\")\n",
- "\n",
- " # Define the model configuration for all components\n",
- " gpt_4o_model = {\n",
- " \"model\": \"gpt-4-turbo-preview\",\n",
- " \"temperature\": 0,\n",
- " \"max_tokens\": 1000,\n",
- " \"top_p\": 1,\n",
- " \"frequency_penalty\": 0,\n",
- " \"presence_penalty\": 0,\n",
- " }\n",
- " print(f\"Component model configuration: {gpt_4o_model}\")\n",
- "\n",
- " try:\n",
- " print(\"Initializing ADAL component...\")\n",
- " adal_component = TrecClassifierAdal(\n",
- " model_client=model_client,\n",
- " model_kwargs=model_kwargs,\n",
- " text_optimizer_model_config=gpt_4o_model,\n",
- " backward_engine_model_config=gpt_4o_model,\n",
- " teacher_model_config=gpt_4o_model,\n",
- " )\n",
- " print(\"ADAL component initialized successfully\")\n",
- "\n",
- " print(\"Initializing trainer...\")\n",
- " trainer = adal.Trainer(\n",
- " train_batch_size=train_batch_size,\n",
- " adaltask=adal_component,\n",
- " strategy=strategy,\n",
- " max_steps=max_steps,\n",
- " num_workers=num_workers,\n",
- " raw_shots=raw_shots,\n",
- " bootstrap_shots=bootstrap_shots,\n",
- " debug=debug,\n",
- " weighted_sampling=True,\n",
- " optimization_order=optimization_order,\n",
- " exclude_input_fields_from_bootstrap_demos=True,\n",
- " )\n",
- " print(\"Trainer initialized successfully\")\n",
- "\n",
- " print(\"Loading datasets...\")\n",
- " train_dataset, val_dataset, test_dataset = load_datasets()\n",
- " print(\n",
- " f\"Datasets loaded - Train size: {len(train_dataset)}, Val size: {len(val_dataset)}, Test size: {len(test_dataset)}\"\n",
- " )\n",
- "\n",
- " print(\"Starting model training...\")\n",
- " trainer.fit(\n",
- " train_dataset=train_dataset,\n",
- " val_dataset=test_dataset,\n",
- " debug=debug,\n",
- " )\n",
- " print(\"Training completed successfully\")\n",
- "\n",
- " except Exception as e:\n",
- " print(f\"Error occurred: {str(e)}\")\n",
- " raise"
- ],
- "metadata": {
- "id": "PEj6xiZ5dVaj"
- },
- "execution_count": 52,
- "outputs": []
- },
- {
- "cell_type": "code",
- "source": [
- "from adalflow.components.model_client.openai_client import OpenAIClient\n",
- "\n",
- "\n",
- "gpt_4o_model = {\n",
- " \"model_client\": OpenAIClient(),\n",
- " \"model_kwargs\": {\n",
- " \"model\": \"gpt-4o-mini\",\n",
- " \"max_tokens\": 2000,\n",
- " },\n",
- "}\n",
- "\n",
- "\n",
- "train(\n",
- " model_client=OpenAIClient(),\n",
- " model_kwargs=gpt_4o_model,\n",
- ")"
- ],
- "metadata": {
- "id": "GnlZBQOMEj6E",
- "collapsed": true
- },
- "execution_count": null,
- "outputs": []
},
- {
- "cell_type": "markdown",
- "source": [
- "# Issues and feedback\n",
- "\n",
- "If you encounter any issues, please report them here: [GitHub Issues](https://github.com/SylphAI-Inc/LightRAG/issues).\n",
- "\n",
- "For feedback, you can use either the [GitHub discussions](https://github.com/SylphAI-Inc/LightRAG/discussions) or [Discord](https://discord.gg/ezzszrRZvT)."
- ],
- "metadata": {
- "id": "AmkbyxmuruUu"
- }
- }
- ]
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "source": [
+ "# ๐ค Welcome to AdalFlow!\n",
+ "## The PyTorch library to auto-optimize any LLM task pipelines\n",
+ "\n",
+ "Thanks for trying us out, we're here to provide you with the best LLM application development experience you can dream of ๐ any questions or concerns you may have, [come talk to us on discord,](https://discord.gg/ezzszrRZvT) we're always here to help! โญ Star us on Github โญ\n",
+ "\n",
+ "\n",
+ "# Quick Links\n",
+ "\n",
+ "Github repo: https://github.com/SylphAI-Inc/AdalFlow\n",
+ "\n",
+ "Full Tutorials: https://adalflow.sylph.ai/index.html#.\n",
+ "\n",
+ "Deep dive on each API: check out the [developer notes](https://adalflow.sylph.ai/tutorials/index.html).\n",
+ "\n",
+ "Common use cases along with the auto-optimization: check out [Use cases](https://adalflow.sylph.ai/use_cases/index.html).\n",
+ "\n",
+ "## ๐ Outline\n",
+ "\n",
+ "This is the code for a classification optimization tutorial \n"
+ ],
+ "metadata": {
+ "id": "xHF95Kr4CzGq"
+ }
+ },
+ {
+ "cell_type": "markdown",
+ "source": [
+ "\n",
+ "# Installation\n",
+ "\n",
+ "1. Use `pip` to install the `adalflow` Python package. We will need `openai`, `groq` from the extra packages.\n",
+ "\n",
+ " ```bash\n",
+ " pip install adalflow[openai,groq]\n",
+ " ```\n",
+ "2. Setup `openai` and `groq` API key in the environment variables\n",
+ "\n",
+ "You can choose to use different client. You can import the model client you prefer. We support `Anthropic`, `Cohere`, `Google`, `GROQ`, `OpenAI`, `Transformer` and more in development. We will use OpenAI here as an example.Please refer to our [full installation guide](https://adalflow.sylph.ai/get_started/installation.html)"
+ ],
+ "metadata": {
+ "id": "Kof5M6DRaKhh"
+ }
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "id": "tAp3eDjOCma1"
+ },
+ "outputs": [],
+ "source": [
+ "from IPython.display import clear_output\n",
+ "\n",
+ "!pip install -U adalflow[openai] # also install the package for the model client you'll use\n",
+ "!pip install datasets\n",
+ "clear_output()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "!pip uninstall httpx anyio -y\n",
+ "!pip install โanyio>=3.1.0,<4.0โ\n",
+ "!pip install httpx==0.24.1"
+ ],
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "id": "CU672Gt4bY7b",
+ "outputId": "532c84d2-c7bd-40ac-c050-e2c5dddc8946"
+ },
+ "execution_count": null,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "Found existing installation: httpx 0.28.1\n",
+ "Uninstalling httpx-0.28.1:\n",
+ " Successfully uninstalled httpx-0.28.1\n",
+ "Found existing installation: anyio 3.7.1\n",
+ "Uninstalling anyio-3.7.1:\n",
+ " Successfully uninstalled anyio-3.7.1\n",
+ "/bin/bash: line 1: 4.0โ: No such file or directory\n",
+ "Collecting httpx==0.24.1\n",
+ " Downloading httpx-0.24.1-py3-none-any.whl.metadata (7.4 kB)\n",
+ "Requirement already satisfied: certifi in /usr/local/lib/python3.10/dist-packages (from httpx==0.24.1) (2024.8.30)\n",
+ "Collecting httpcore<0.18.0,>=0.15.0 (from httpx==0.24.1)\n",
+ " Downloading httpcore-0.17.3-py3-none-any.whl.metadata (18 kB)\n",
+ "Requirement already satisfied: idna in /usr/local/lib/python3.10/dist-packages (from httpx==0.24.1) (3.10)\n",
+ "Requirement already satisfied: sniffio in /usr/local/lib/python3.10/dist-packages (from httpx==0.24.1) (1.3.1)\n",
+ "Requirement already satisfied: h11<0.15,>=0.13 in /usr/local/lib/python3.10/dist-packages (from httpcore<0.18.0,>=0.15.0->httpx==0.24.1) (0.14.0)\n",
+ "Collecting anyio<5.0,>=3.0 (from httpcore<0.18.0,>=0.15.0->httpx==0.24.1)\n",
+ " Downloading anyio-4.7.0-py3-none-any.whl.metadata (4.7 kB)\n",
+ "Requirement already satisfied: exceptiongroup>=1.0.2 in /usr/local/lib/python3.10/dist-packages (from anyio<5.0,>=3.0->httpcore<0.18.0,>=0.15.0->httpx==0.24.1) (1.2.2)\n",
+ "Requirement already satisfied: typing_extensions>=4.5 in /usr/local/lib/python3.10/dist-packages (from anyio<5.0,>=3.0->httpcore<0.18.0,>=0.15.0->httpx==0.24.1) (4.12.2)\n",
+ "Downloading httpx-0.24.1-py3-none-any.whl (75 kB)\n",
+ "\u001b[2K \u001b[90mโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\u001b[0m \u001b[32m75.4/75.4 kB\u001b[0m \u001b[31m2.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
+ "\u001b[?25hDownloading httpcore-0.17.3-py3-none-any.whl (74 kB)\n",
+ "\u001b[2K \u001b[90mโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\u001b[0m \u001b[32m74.5/74.5 kB\u001b[0m \u001b[31m6.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
+ "\u001b[?25hDownloading anyio-4.7.0-py3-none-any.whl (93 kB)\n",
+ "\u001b[2K \u001b[90mโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\u001b[0m \u001b[32m93.1/93.1 kB\u001b[0m \u001b[31m8.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
+ "\u001b[?25hInstalling collected packages: anyio, httpcore, httpx\n",
+ " Attempting uninstall: httpcore\n",
+ " Found existing installation: httpcore 1.0.7\n",
+ " Uninstalling httpcore-1.0.7:\n",
+ " Successfully uninstalled httpcore-1.0.7\n",
+ "\u001b[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n",
+ "jupyter-server 1.24.0 requires anyio<4,>=3.1.0, but you have anyio 4.7.0 which is incompatible.\u001b[0m\u001b[31m\n",
+ "\u001b[0mSuccessfully installed anyio-4.7.0 httpcore-0.17.3 httpx-0.24.1\n"
+ ]
+ }
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "source": [
+ "## Set Environment Variables\n",
+ "\n",
+ "Run the following code and pass your api key.\n",
+ "\n",
+ "Note: for normal `.py` projects, follow our [official installation guide](https://lightrag.sylph.ai/get_started/installation.html).\n",
+ "\n",
+ "*Go to [OpenAI](https://platform.openai.com/docs/introduction) to get API keys if you don't already have.*"
+ ],
+ "metadata": {
+ "id": "KapUyHMM07pJ"
+ }
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "import os\n",
+ "\n",
+ "from getpass import getpass\n",
+ "\n",
+ "# Prompt user to enter their API keys securely\n",
+ "openai_api_key = getpass(\"Please enter your OpenAI API key: \")\n",
+ "\n",
+ "\n",
+ "# Set environment variables\n",
+ "os.environ[\"OPENAI_API_KEY\"] = openai_api_key\n",
+ "\n",
+ "print(\"API keys have been set.\")"
+ ],
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "id": "ONfzF9Puzdd_",
+ "outputId": "a8ca0388-be6e-4b7a-cd05-d4ec52f64e95"
+ },
+ "execution_count": null,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "Please enter your OpenAI API key: ยทยทยทยทยทยทยทยทยทยท\n",
+ "API keys have been set.\n"
+ ]
+ }
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "source": [
+ "Prepare data structures and prompt template"
+ ],
+ "metadata": {
+ "id": "4W3yEpRpepNK"
+ }
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "from dataclasses import dataclass, field\n",
+ "from typing import List, Dict, Union, Optional, Tuple, Any, Callable\n",
+ "from datasets import load_dataset\n",
+ "from adalflow.components.model_client import OpenAIClient\n",
+ "import adalflow as adal\n",
+ "from adalflow.core.component import Component\n",
+ "from adalflow.datasets.types import TrecData\n",
+ "from adalflow.datasets.trec import TrecDataset\n",
+ "\n",
+ "from adalflow.eval.answer_match_acc import AnswerMatchAcc\n",
+ "\n",
+ "\n",
+ "_COARSE_LABELS = [\"ABBR\", \"DESC\", \"ENTY\", \"HUM\", \"LOC\", \"NUM\"]\n",
+ "\n",
+ "_COARSE_LABELS_DESC = [\n",
+ " \"Abbreviation: Questions about abbreviations and their meanings\",\n",
+ " \"Description: Questions seeking descriptions of people, things, or concepts\",\n",
+ " \"Entity: Questions about entities (e.g., animals, colors, inventions)\",\n",
+ " \"Human: Questions about people or organizations\",\n",
+ " \"Location: Questions about places, cities, countries\",\n",
+ " \"Numeric: Questions seeking numeric answers (e.g., dates, amounts, distances)\",\n",
+ "]\n",
+ "\n",
+ "\n",
+ "template = r\"\"\"\n",
+ " {{system_prompt}}\n",
+ " {% if output_format_str is not none %}\n",
+ " {{output_format_str}}\n",
+ " {% endif %}\n",
+ " {% if few_shot_demos is not none %}\n",
+ " Here are some examples:\n",
+ " {{few_shot_demos}}\n",
+ " {% endif %}\n",
+ " \n",
+ " \n",
+ " {{input_str}}\n",
+ " \n",
+ " \"\"\"\n",
+ "\n",
+ "task_desc_template = r\"\"\"You are a classifier. Given a question, you need to classify it into one of the following classes:\n",
+ " Format: class_index. class_name, class_description\n",
+ " {% if classes %}\n",
+ " {% for class in classes %}\n",
+ " {{loop.index-1}}. {{class.label}}, {{class.desc}}\n",
+ " {% endfor %}\n",
+ " {% endif %}\n",
+ " - Do not try to answer the question:\n",
+ " \"\"\"\n",
+ "\n",
+ "\n",
+ "@dataclass\n",
+ "class TRECExtendedData(TrecData):\n",
+ " rationale: str = field(\n",
+ " metadata={\n",
+ " \"desc\": \"Your step-by-step reasoning to classify the question to class_name\"\n",
+ " },\n",
+ " default=None,\n",
+ " )\n",
+ " __input_fields__ = [\"question\"]\n",
+ " __output_fields__ = [\n",
+ " \"rationale\",\n",
+ " \"class_name\",\n",
+ " ] # it is important to have the rationale before the class_name\n",
+ "\n",
+ "def load_datasets():\n",
+ " \"\"\"Load the dataset\"\"\"\n",
+ " train_data = TrecDataset(split=\"train\")\n",
+ " val_data = TrecDataset(split=\"val\")\n",
+ " test_data = TrecDataset(split=\"test\")\n",
+ " return train_data, val_data, test_data # 0.694, 0.847"
+ ],
+ "metadata": {
+ "id": "ZZIEtZYHNVjo"
+ },
+ "execution_count": null,
+ "outputs": []
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "# prepare models\n",
+ "\n",
+ "from adalflow.components.model_client.openai_client import OpenAIClient\n",
+ "\n",
+ "# used as the target model\n",
+ "gpt_3_model = {\n",
+ " \"model_client\": OpenAIClient(),\n",
+ " \"model_kwargs\": {\n",
+ " \"model\": \"gpt-3.5-turbo\",\n",
+ " \"max_tokens\": 2000,\n",
+ " \"temperature\": 0.0,\n",
+ " \"top_p\": 0.99,\n",
+ " \"frequency_penalty\": 0,\n",
+ " \"presence_penalty\": 0,\n",
+ " \"stop\": None,\n",
+ " },\n",
+ "}\n",
+ "\n",
+ "# used as optimizer and backward engine\n",
+ "gpt_4o_mini_model = {\n",
+ " \"model_client\": OpenAIClient(),\n",
+ " \"model_kwargs\": {\n",
+ " \"model\": \"gpt-4o-mini\",\n",
+ " \"temperature\": 1,\n",
+ " \"top_p\": 0.99,\n",
+ " \"max_tokens\": 1000,\n",
+ " # \"frequency_penalty\": 1, # high for nto repeating prompt\n",
+ " },\n",
+ "}"
+ ],
+ "metadata": {
+ "id": "yAvzn7DZeUX-"
+ },
+ "execution_count": null,
+ "outputs": []
+ },
+ {
+ "cell_type": "markdown",
+ "source": [
+ "Create the task pipeline"
+ ],
+ "metadata": {
+ "id": "G664uy9MgDdC"
+ }
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "class TRECClassifierStructuredOutput(adal.Component):\n",
+ "\n",
+ " def __init__(self, model_client: adal.ModelClient, model_kwargs: Dict):\n",
+ " super().__init__()\n",
+ "\n",
+ " label_desc = [\n",
+ " {\"label\": label, \"desc\": desc}\n",
+ " for label, desc in zip(_COARSE_LABELS, _COARSE_LABELS_DESC)\n",
+ " ]\n",
+ "\n",
+ " task_desc_str = adal.Prompt(\n",
+ " template=task_desc_template, prompt_kwargs={\"classes\": label_desc}\n",
+ " )()\n",
+ "\n",
+ " self.data_class = TRECExtendedData\n",
+ " self.data_class.set_task_desc(task_desc_str)\n",
+ "\n",
+ " self.parser = adal.DataClassParser(\n",
+ " data_class=self.data_class, return_data_class=True, format_type=\"yaml\"\n",
+ " )\n",
+ "\n",
+ " prompt_kwargs = {\n",
+ " \"system_prompt\": adal.Parameter(\n",
+ " data=self.parser.get_task_desc_str(),\n",
+ " role_desc=\"Task description\",\n",
+ " requires_opt=True,\n",
+ " param_type=adal.ParameterType.PROMPT,\n",
+ " ),\n",
+ " \"output_format_str\": adal.Parameter(\n",
+ " data=self.parser.get_output_format_str(),\n",
+ " role_desc=\"Output format requirements\",\n",
+ " requires_opt=False,\n",
+ " param_type=adal.ParameterType.PROMPT,\n",
+ " ),\n",
+ " \"few_shot_demos\": adal.Parameter(\n",
+ " data=None,\n",
+ " requires_opt=True,\n",
+ " role_desc=\"Few shot examples to help the model\",\n",
+ " param_type=adal.ParameterType.DEMOS,\n",
+ " ),\n",
+ " }\n",
+ "\n",
+ " self.llm = adal.Generator(\n",
+ " model_client=model_client,\n",
+ " model_kwargs=model_kwargs,\n",
+ " prompt_kwargs=prompt_kwargs,\n",
+ " template=template,\n",
+ " output_processors=self.parser,\n",
+ " use_cache=True,\n",
+ " )\n",
+ "\n",
+ " def _prepare_input(self, question: str):\n",
+ " input_data = self.data_class(question=question)\n",
+ " input_str = self.parser.get_input_str(input_data)\n",
+ " prompt_kwargs = {\n",
+ " \"input_str\": adal.Parameter(\n",
+ " data=input_str, requires_opt=False, role_desc=\"input to the LLM\"\n",
+ " )\n",
+ " }\n",
+ " return prompt_kwargs\n",
+ "\n",
+ " def call(\n",
+ " self, question: str, id: Optional[str] = None\n",
+ " ) -> Union[adal.GeneratorOutput, adal.Parameter]:\n",
+ " prompt_kwargs = self._prepare_input(question)\n",
+ " output = self.llm(prompt_kwargs=prompt_kwargs, id=id)\n",
+ " return output"
+ ],
+ "metadata": {
+ "id": "3Q3H9XC4Ncfi"
+ },
+ "execution_count": null,
+ "outputs": []
+ },
+ {
+ "cell_type": "markdown",
+ "source": [
+ "Inference the task pipeline and draw the computation graph"
+ ],
+ "metadata": {
+ "id": "gj08oOqqgGyr"
+ }
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "# load dataset to get one example\n",
+ "\n",
+ "train_dataset, val_dataset, test_dataset = load_datasets()\n",
+ "example = train_dataset[0]\n",
+ "print(example)"
+ ],
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "id": "qtvLN8zOgnSg",
+ "outputId": "9996f8c3-371d-4b5c-ec48-e8cf6d6c396b"
+ },
+ "execution_count": null,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "TrecData(id='e73a82a7-6a3d-4947-90f5-03739e169db0', question='When reading classified ads , what does EENTY : other stand for ?', class_name='ABBR', class_index=0)\n"
+ ]
+ }
+ ]
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "task = TRECClassifierStructuredOutput(\n",
+ " model_client=gpt_3_model[\"model_client\"],\n",
+ " model_kwargs=gpt_3_model[\"model_kwargs\"],\n",
+ ")\n",
+ "task.train()\n",
+ "\n",
+ "output = task(question=example.question, id=example.id)\n",
+ "print(output)"
+ ],
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "id": "cKuW3QlhgLTG",
+ "outputId": "7f1f9cd6-9615-4b41-ecc5-5901626d57ae"
+ },
+ "execution_count": null,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "Parameter(name=Generator_output, requires_opt=True, param_type=generator_output (The output of the generator.), role_desc=Output from (llm) Generator, data=```\n",
+ "rationale: The question is asking for the meaning of the abbreviation \"EENTY\" in classified ads, which falls under the ABBR class.\n",
+ "class_name: ABBR\n",
+ "```, predecessors={Parameter(name=Output_for, requires_opt=False, param_type=prompt (Instruction to the language model on task, data, and format.), role_desc=Output format requirements, data=Your output should be formatted as a standard YAML instance with the following schema:\n",
+ "```\n",
+ "rationale: Your step-by-step reasoning to classify the question to class_name (str) (optional)\n",
+ "class_name: One of {ABBR, ENTY, DESC, HUM, LOC, NUM} (str) (optional)\n",
+ "```\n",
+ "-Make sure to always enclose the YAML output in triple backticks (```). Please do not add anything other than valid YAML output!\n",
+ "-Follow the YAML formatting conventions with an indent of 2 spaces.\n",
+ "-DO NOT mistaken the \"properties\" and \"type\" in the schema as the actual fields in the YAML output.\n",
+ "-Quote the string values properly., predecessors=set(), gradients=[], raw_response=None, input_args=None, traces={}), Parameter(name=Few_shot_e, requires_opt=True, param_type=demos (A few examples to guide the language model.), role_desc=Few shot examples to help the model, data=None, predecessors=set(), gradients=[], raw_response=None, input_args=None, traces={}), Parameter(name=Input_to_t, requires_opt=False, param_type=none (), role_desc=input to the LLM, data=question: 'When reading classified ads , what does EENTY : other stand for ?', predecessors=set(), gradients=[], raw_response=None, input_args=None, traces={}), Parameter(name=Task_descr, requires_opt=True, param_type=prompt (Instruction to the language model on task, data, and format.), role_desc=Task description, data=You are a classifier. Given a question, you need to classify it into one of the following classes:\n",
+ " Format: class_index. class_name, class_description\n",
+ " 0. ABBR, Abbreviation: Questions about abbreviations and their meanings\n",
+ " 1. DESC, Description: Questions seeking descriptions of people, things, or concepts\n",
+ " 2. ENTY, Entity: Questions about entities (e.g., animals, colors, inventions)\n",
+ " 3. HUM, Human: Questions about people or organizations\n",
+ " 4. LOC, Location: Questions about places, cities, countries\n",
+ " 5. NUM, Numeric: Questions seeking numeric answers (e.g., dates, amounts, distances)\n",
+ " - Do not try to answer the question:\n",
+ " , predecessors=set(), gradients=[], raw_response=None, input_args=None, traces={})}, gradients=[], raw_response=None, input_args={'prompt_kwargs': {'system_prompt': Parameter(name=Task_descr, requires_opt=True, param_type=prompt (Instruction to the language model on task, data, and format.), role_desc=Task description, data=You are a classifier. Given a question, you need to classify it into one of the following classes:\n",
+ " Format: class_index. class_name, class_description\n",
+ " 0. ABBR, Abbreviation: Questions about abbreviations and their meanings\n",
+ " 1. DESC, Description: Questions seeking descriptions of people, things, or concepts\n",
+ " 2. ENTY, Entity: Questions about entities (e.g., animals, colors, inventions)\n",
+ " 3. HUM, Human: Questions about people or organizations\n",
+ " 4. LOC, Location: Questions about places, cities, countries\n",
+ " 5. NUM, Numeric: Questions seeking numeric answers (e.g., dates, amounts, distances)\n",
+ " - Do not try to answer the question:\n",
+ " , predecessors=set(), gradients=[], raw_response=None, input_args=None, traces={}), 'output_format_str': Parameter(name=Output_for, requires_opt=False, param_type=prompt (Instruction to the language model on task, data, and format.), role_desc=Output format requirements, data=Your output should be formatted as a standard YAML instance with the following schema:\n",
+ "```\n",
+ "rationale: Your step-by-step reasoning to classify the question to class_name (str) (optional)\n",
+ "class_name: One of {ABBR, ENTY, DESC, HUM, LOC, NUM} (str) (optional)\n",
+ "```\n",
+ "-Make sure to always enclose the YAML output in triple backticks (```). Please do not add anything other than valid YAML output!\n",
+ "-Follow the YAML formatting conventions with an indent of 2 spaces.\n",
+ "-DO NOT mistaken the \"properties\" and \"type\" in the schema as the actual fields in the YAML output.\n",
+ "-Quote the string values properly., predecessors=set(), gradients=[], raw_response=None, input_args=None, traces={}), 'few_shot_demos': Parameter(name=Few_shot_e, requires_opt=True, param_type=demos (A few examples to guide the language model.), role_desc=Few shot examples to help the model, data=None, predecessors=set(), gradients=[], raw_response=None, input_args=None, traces={}), 'input_str': Parameter(name=Input_to_t, requires_opt=False, param_type=none (), role_desc=input to the LLM, data=question: 'When reading classified ads , what does EENTY : other stand for ?', predecessors=set(), gradients=[], raw_response=None, input_args=None, traces={})}, 'model_kwargs': {'model': 'gpt-3.5-turbo', 'max_tokens': 2000, 'temperature': 0.0, 'top_p': 0.99, 'frequency_penalty': 0, 'presence_penalty': 0, 'stop': None}}, traces={})\n"
+ ]
+ }
+ ]
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "class TrecClassifierAdal(adal.AdalComponent):\n",
+ " def __init__(\n",
+ " self,\n",
+ " model_client: adal.ModelClient,\n",
+ " model_kwargs: Dict,\n",
+ " teacher_model_config: Dict,\n",
+ " backward_engine_model_config: Dict,\n",
+ " text_optimizer_model_config: Dict,\n",
+ " ):\n",
+ " task = TRECClassifierStructuredOutput(model_client, model_kwargs)\n",
+ " eval_fn = AnswerMatchAcc(type=\"exact_match\").compute_single_item\n",
+ " loss_fn = adal.EvalFnToTextLoss(\n",
+ " eval_fn=eval_fn,\n",
+ " eval_fn_desc=\"exact_match: 1 if str(y) == str(y_gt) else 0\",\n",
+ " )\n",
+ " super().__init__(\n",
+ " task=task,\n",
+ " eval_fn=eval_fn,\n",
+ " loss_fn=loss_fn,\n",
+ " backward_engine_model_config=backward_engine_model_config,\n",
+ " text_optimizer_model_config=text_optimizer_model_config,\n",
+ " teacher_model_config=teacher_model_config,\n",
+ " )\n",
+ "\n",
+ " def prepare_task(self, sample: TRECExtendedData):\n",
+ " return self.task.call, {\"question\": sample.question, \"id\": sample.id}\n",
+ "\n",
+ " def prepare_eval(\n",
+ " self, sample: TRECExtendedData, y_pred: adal.GeneratorOutput\n",
+ " ) -> float:\n",
+ " y_label = -1\n",
+ " if y_pred and y_pred.data is not None and y_pred.data.class_name is not None:\n",
+ " y_label = y_pred.data.class_name\n",
+ " return self.eval_fn, {\"y\": y_label, \"y_gt\": sample.class_name}\n",
+ "\n",
+ " def prepare_loss(\n",
+ " self, sample: TRECExtendedData, y_pred: adal.Parameter, *args, **kwargs\n",
+ " ) -> Tuple[Callable[..., Any], Dict]:\n",
+ " full_response = y_pred.full_response\n",
+ " y_label = -1\n",
+ " if (\n",
+ " full_response\n",
+ " and full_response.data is not None\n",
+ " and full_response.data.class_name is not None\n",
+ " ):\n",
+ " y_label = full_response.data.class_name\n",
+ "\n",
+ " y_pred.eval_input = y_label\n",
+ " y_gt = adal.Parameter(\n",
+ " name=\"y_gt\",\n",
+ " data=sample.class_name,\n",
+ " eval_input=sample.class_name,\n",
+ " requires_opt=False,\n",
+ " )\n",
+ " return self.loss_fn, {\"kwargs\": {\"y\": y_pred, \"y_gt\": y_gt}}"
+ ],
+ "metadata": {
+ "id": "HpkQYsh2NevT"
+ },
+ "execution_count": null,
+ "outputs": []
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "def train(\n",
+ " model_client: adal.ModelClient,\n",
+ " model_kwargs: Dict,\n",
+ " train_batch_size=4,\n",
+ " raw_shots: int = 0,\n",
+ " bootstrap_shots: int = 1,\n",
+ " max_steps=12,\n",
+ " num_workers=4,\n",
+ " strategy=\"constrained\",\n",
+ " optimization_order=\"sequential\",\n",
+ " debug=False,\n",
+ "):\n",
+ " print(\"Starting training process...\")\n",
+ "\n",
+ " # Define the model configuration for all components\n",
+ " gpt_4o_model = {\n",
+ " \"model_client\": OpenAIClient(),\n",
+ " \"model_kwargs\": {\n",
+ " \"model\": \"gpt-4o-mini\",\n",
+ " \"temperature\": 1,\n",
+ " \"top_p\": 0.99,\n",
+ " \"max_tokens\": 1000,\n",
+ " # \"frequency_penalty\": 1, # high for nto repeating prompt\n",
+ " },\n",
+ " }\n",
+ "\n",
+ " print(f\"Component model configuration: {gpt_4o_model}\")\n",
+ "\n",
+ " try:\n",
+ " print(\"Initializing ADAL component...\")\n",
+ " adal_component = TrecClassifierAdal(\n",
+ " model_client=model_client,\n",
+ " model_kwargs=model_kwargs,\n",
+ " text_optimizer_model_config=gpt_4o_model,\n",
+ " backward_engine_model_config=gpt_4o_model,\n",
+ " teacher_model_config=gpt_4o_model,\n",
+ " )\n",
+ " print(\"ADAL component initialized successfully\")\n",
+ "\n",
+ " print(\"Initializing trainer...\")\n",
+ " trainer = adal.Trainer(\n",
+ " train_batch_size=train_batch_size,\n",
+ " adaltask=adal_component,\n",
+ " strategy=strategy,\n",
+ " max_steps=max_steps,\n",
+ " num_workers=num_workers,\n",
+ " raw_shots=raw_shots,\n",
+ " bootstrap_shots=bootstrap_shots,\n",
+ " debug=debug,\n",
+ " weighted_sampling=True,\n",
+ " optimization_order=optimization_order,\n",
+ " exclude_input_fields_from_bootstrap_demos=True,\n",
+ " )\n",
+ " print(\"Trainer initialized successfully\")\n",
+ "\n",
+ " print(\"Loading datasets...\")\n",
+ " train_dataset, val_dataset, test_dataset = load_datasets()\n",
+ " print(\n",
+ " f\"Datasets loaded - Train size: {len(train_dataset)}, Val size: {len(val_dataset)}, Test size: {len(test_dataset)}\"\n",
+ " )\n",
+ "\n",
+ " print(\"Starting model training...\")\n",
+ " trainer.fit(\n",
+ " train_dataset=train_dataset,\n",
+ " val_dataset=test_dataset,\n",
+ " debug=debug,\n",
+ " )\n",
+ " print(\"Training completed successfully\")\n",
+ "\n",
+ " except Exception as e:\n",
+ " print(f\"Error occurred: {str(e)}\")\n",
+ " raise"
+ ],
+ "metadata": {
+ "id": "PEj6xiZ5dVaj"
+ },
+ "execution_count": null,
+ "outputs": []
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "train(**gpt_3_model)"
+ ],
+ "metadata": {
+ "id": "GnlZBQOMEj6E",
+ "colab": {
+ "base_uri": "https://localhost:8080/",
+ "height": 1000
+ },
+ "outputId": "055a95c4-ccae-4028-d904-86b839bc1c14"
+ },
+ "execution_count": null,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "Starting training process...\n",
+ "Component model configuration: {'model_client': OpenAIClient(), 'model_kwargs': {'model': 'gpt-4o-mini', 'temperature': 1, 'top_p': 0.99, 'max_tokens': 1000}}\n",
+ "Initializing ADAL component...\n",
+ "ADAL component initialized successfully\n",
+ "Initializing trainer...\n",
+ "Trainer initialized successfully\n",
+ "Loading datasets...\n",
+ "Datasets loaded - Train size: 120, Val size: 36, Test size: 144\n",
+ "Starting model training...\n",
+ "raw_shots: 0, bootstrap_shots: 1\n",
+ "Configuring teacher generator.\n",
+ "Configuring teacher generator for Generator(\n",
+ " model_kwargs={'model': 'gpt-4o-mini', 'temperature': 1, 'top_p': 0.99, 'max_tokens': 1000}, trainable_prompt_kwargs=[]\n",
+ " (prompt): Prompt(\n",
+ " template: \n",
+ " {{system_prompt}}\n",
+ " {% if output_format_str is not none %}\n",
+ " {{output_format_str}}\n",
+ " {% endif %}\n",
+ " {% if few_shot_demos is not none %}\n",
+ " Here are some examples:\n",
+ " {{few_shot_demos}}\n",
+ " {% endif %}\n",
+ " \n",
+ " \n",
+ " {{input_str}}\n",
+ " \n",
+ " , prompt_kwargs: {'system_prompt': 'You are a classifier. Given a question, you need to classify it into one of the following classes:\\n Format: class_index. class_name, class_description\\n 0. ABBR, Abbreviation: Questions about abbreviations and their meanings\\n 1. DESC, Description: Questions seeking descriptions of people, things, or concepts\\n 2. ENTY, Entity: Questions about entities (e.g., animals, colors, inventions)\\n 3. HUM, Human: Questions about people or organizations\\n 4. LOC, Location: Questions about places, cities, countries\\n 5. NUM, Numeric: Questions seeking numeric answers (e.g., dates, amounts, distances)\\n - Do not try to answer the question:\\n ', 'output_format_str': 'Your output should be formatted as a standard YAML instance with the following schema:\\n```\\nrationale: Your step-by-step reasoning to classify the question to class_name (str) (optional)\\nclass_name: One of {ABBR, ENTY, DESC, HUM, LOC, NUM} (str) (optional)\\n```\\n-Make sure to always enclose the YAML output in triple backticks (```). Please do not add anything other than valid YAML output!\\n-Follow the YAML formatting conventions with an indent of 2 spaces.\\n-DO NOT mistaken the \"properties\" and \"type\" in the schema as the actual fields in the YAML output.\\n-Quote the string values properly.', 'few_shot_demos': 'None'}, prompt_variables: ['output_format_str', 'system_prompt', 'input_str', 'few_shot_demos']\n",
+ " )\n",
+ " (model_client): OpenAIClient()\n",
+ " (output_processors): DataClassParser(\n",
+ " data_class=TRECExtendedData, format_type=yaml, return_data_class=True, input_fields=['question'], output_fields=['rationale', 'class_name']\n",
+ " (_output_processor): YamlParser()\n",
+ " (output_format_prompt): Prompt(\n",
+ " template: Your output should be formatted as a standard YAML instance with the following schema:\n",
+ " ```\n",
+ " {{schema}}\n",
+ " ```\n",
+ " -Make sure to always enclose the YAML output in triple backticks (```). Please do not add anything other than valid YAML output!\n",
+ " -Follow the YAML formatting conventions with an indent of 2 spaces.\n",
+ " -DO NOT mistaken the \"properties\" and \"type\" in the schema as the actual fields in the YAML output.\n",
+ " -Quote the string values properly., prompt_variables: ['schema']\n",
+ " )\n",
+ " )\n",
+ ")\n",
+ "Teacher generator set: Generator(\n",
+ " model_kwargs={'model': 'gpt-4o-mini', 'temperature': 1, 'top_p': 0.99, 'max_tokens': 1000}, trainable_prompt_kwargs=[]\n",
+ " (prompt): Prompt(\n",
+ " template: \n",
+ " {{system_prompt}}\n",
+ " {% if output_format_str is not none %}\n",
+ " {{output_format_str}}\n",
+ " {% endif %}\n",
+ " {% if few_shot_demos is not none %}\n",
+ " Here are some examples:\n",
+ " {{few_shot_demos}}\n",
+ " {% endif %}\n",
+ " \n",
+ " \n",
+ " {{input_str}}\n",
+ " \n",
+ " , prompt_kwargs: {'system_prompt': 'You are a classifier. Given a question, you need to classify it into one of the following classes:\\n Format: class_index. class_name, class_description\\n 0. ABBR, Abbreviation: Questions about abbreviations and their meanings\\n 1. DESC, Description: Questions seeking descriptions of people, things, or concepts\\n 2. ENTY, Entity: Questions about entities (e.g., animals, colors, inventions)\\n 3. HUM, Human: Questions about people or organizations\\n 4. LOC, Location: Questions about places, cities, countries\\n 5. NUM, Numeric: Questions seeking numeric answers (e.g., dates, amounts, distances)\\n - Do not try to answer the question:\\n ', 'output_format_str': 'Your output should be formatted as a standard YAML instance with the following schema:\\n```\\nrationale: Your step-by-step reasoning to classify the question to class_name (str) (optional)\\nclass_name: One of {ABBR, ENTY, DESC, HUM, LOC, NUM} (str) (optional)\\n```\\n-Make sure to always enclose the YAML output in triple backticks (```). Please do not add anything other than valid YAML output!\\n-Follow the YAML formatting conventions with an indent of 2 spaces.\\n-DO NOT mistaken the \"properties\" and \"type\" in the schema as the actual fields in the YAML output.\\n-Quote the string values properly.', 'few_shot_demos': 'None'}, prompt_variables: ['output_format_str', 'system_prompt', 'input_str', 'few_shot_demos']\n",
+ " )\n",
+ " (model_client): OpenAIClient()\n",
+ " (output_processors): DataClassParser(\n",
+ " data_class=TRECExtendedData, format_type=yaml, return_data_class=True, input_fields=['question'], output_fields=['rationale', 'class_name']\n",
+ " (_output_processor): YamlParser()\n",
+ " (output_format_prompt): Prompt(\n",
+ " template: Your output should be formatted as a standard YAML instance with the following schema:\n",
+ " ```\n",
+ " {{schema}}\n",
+ " ```\n",
+ " -Make sure to always enclose the YAML output in triple backticks (```). Please do not add anything other than valid YAML output!\n",
+ " -Follow the YAML formatting conventions with an indent of 2 spaces.\n",
+ " -DO NOT mistaken the \"properties\" and \"type\" in the schema as the actual fields in the YAML output.\n",
+ " -Quote the string values properly., prompt_variables: ['schema']\n",
+ " )\n",
+ " )\n",
+ "), teacher Generator(\n",
+ " model_kwargs={'model': 'gpt-4o-mini', 'temperature': 1, 'top_p': 0.99, 'max_tokens': 1000}, trainable_prompt_kwargs=[]\n",
+ " (prompt): Prompt(\n",
+ " template: \n",
+ " {{system_prompt}}\n",
+ " {% if output_format_str is not none %}\n",
+ " {{output_format_str}}\n",
+ " {% endif %}\n",
+ " {% if few_shot_demos is not none %}\n",
+ " Here are some examples:\n",
+ " {{few_shot_demos}}\n",
+ " {% endif %}\n",
+ " \n",
+ " \n",
+ " {{input_str}}\n",
+ " \n",
+ " , prompt_kwargs: {'system_prompt': 'You are a classifier. Given a question, you need to classify it into one of the following classes:\\n Format: class_index. class_name, class_description\\n 0. ABBR, Abbreviation: Questions about abbreviations and their meanings\\n 1. DESC, Description: Questions seeking descriptions of people, things, or concepts\\n 2. ENTY, Entity: Questions about entities (e.g., animals, colors, inventions)\\n 3. HUM, Human: Questions about people or organizations\\n 4. LOC, Location: Questions about places, cities, countries\\n 5. NUM, Numeric: Questions seeking numeric answers (e.g., dates, amounts, distances)\\n - Do not try to answer the question:\\n ', 'output_format_str': 'Your output should be formatted as a standard YAML instance with the following schema:\\n```\\nrationale: Your step-by-step reasoning to classify the question to class_name (str) (optional)\\nclass_name: One of {ABBR, ENTY, DESC, HUM, LOC, NUM} (str) (optional)\\n```\\n-Make sure to always enclose the YAML output in triple backticks (```). Please do not add anything other than valid YAML output!\\n-Follow the YAML formatting conventions with an indent of 2 spaces.\\n-DO NOT mistaken the \"properties\" and \"type\" in the schema as the actual fields in the YAML output.\\n-Quote the string values properly.', 'few_shot_demos': 'None'}, prompt_variables: ['output_format_str', 'system_prompt', 'input_str', 'few_shot_demos']\n",
+ " )\n",
+ " (model_client): OpenAIClient()\n",
+ " (output_processors): DataClassParser(\n",
+ " data_class=TRECExtendedData, format_type=yaml, return_data_class=True, input_fields=['question'], output_fields=['rationale', 'class_name']\n",
+ " (_output_processor): YamlParser()\n",
+ " (output_format_prompt): Prompt(\n",
+ " template: Your output should be formatted as a standard YAML instance with the following schema:\n",
+ " ```\n",
+ " {{schema}}\n",
+ " ```\n",
+ " -Make sure to always enclose the YAML output in triple backticks (```). Please do not add anything other than valid YAML output!\n",
+ " -Follow the YAML formatting conventions with an indent of 2 spaces.\n",
+ " -DO NOT mistaken the \"properties\" and \"type\" in the schema as the actual fields in the YAML output.\n",
+ " -Quote the string values properly., prompt_variables: ['schema']\n",
+ " )\n",
+ " )\n",
+ ")\n",
+ "Teacher generator configured.\n",
+ "Configured demo optimizers\n",
+ "Backward engine configured for all generators.\n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stderr",
+ "text": [
+ "\n",
+ "Loading Data: 100%|โโโโโโโโโโ| 144/144 [00:00<00:00, 9161.62it/s]\n",
+ "Predicting: step(0): 0.8264 across 144 samples, Max potential: 0.8264: 100%|โโโโโโโโโโ| 144/144 [00:19<00:00, 7.39it/s]\n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "completed_samples: 144, len: 144\n",
+ "Initial validation score: 0.8263888888888888\n",
+ "Initial test score: None\n",
+ "Checkpoint path: /root/.adalflow/ckpt/TrecClassifierAdal\n",
+ "save to /root/.adalflow/ckpt/TrecClassifierAdal/constrained_max_steps_12_a6e76_run_1.json\n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stderr",
+ "text": [
+ "\n",
+ "Training Step: 1: 0%| | 0/30 [00:00\u001b[0m in \u001b[0;36m\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mtrain\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0mgpt_3_model\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
+ "\u001b[0;32m\u001b[0m in \u001b[0;36mtrain\u001b[0;34m(model_client, model_kwargs, train_batch_size, raw_shots, bootstrap_shots, max_steps, num_workers, strategy, optimization_order, debug)\u001b[0m\n\u001b[1;32m 61\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 62\u001b[0m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Starting model training...\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 63\u001b[0;31m trainer.fit(\n\u001b[0m\u001b[1;32m 64\u001b[0m \u001b[0mtrain_dataset\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtrain_dataset\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 65\u001b[0m \u001b[0mval_dataset\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtest_dataset\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/adalflow/optim/trainer/trainer.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, adaltask, train_loader, train_dataset, val_dataset, test_dataset, debug, save_traces, raw_shots, bootstrap_shots, resume_from_ckpt)\u001b[0m\n\u001b[1;32m 477\u001b[0m \u001b[0mstarting_step\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmax_steps\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 478\u001b[0m \u001b[0;32melif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstrategy\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m\"constrained\"\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 479\u001b[0;31m trainer_results = self._fit_text_grad_constraint(\n\u001b[0m\u001b[1;32m 480\u001b[0m \u001b[0mtrain_loader\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 481\u001b[0m \u001b[0mval_dataset\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/adalflow/optim/trainer/trainer.py\u001b[0m in \u001b[0;36m_fit_text_grad_constraint\u001b[0;34m(self, train_loader, val_dataset, test_dataset, trainer_results, starting_step)\u001b[0m\n\u001b[1;32m 1779\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1780\u001b[0m all_samples, all_losses, all_y_preds = (\n\u001b[0;32m-> 1781\u001b[0;31m self._text_grad_constraint_propose_step(\n\u001b[0m\u001b[1;32m 1782\u001b[0m \u001b[0msteps\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0msteps\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1783\u001b[0m \u001b[0mall_samples\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mall_samples\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/adalflow/optim/trainer/trainer.py\u001b[0m in \u001b[0;36m_text_grad_constraint_propose_step\u001b[0;34m(self, steps, all_samples, all_losses, all_y_preds, include_demo_optimizers)\u001b[0m\n\u001b[1;32m 1657\u001b[0m \u001b[0;31m# print(f\"Proposing step: {i}\")\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1658\u001b[0m \u001b[0;31m# self.optimizer.propose()\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1659\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_propose_text_optimizers\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;31m# new prompts\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1660\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0minclude_demo_optimizers\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1661\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_demo_optimizers_propose\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/adalflow/optim/trainer/trainer.py\u001b[0m in \u001b[0;36m_propose_text_optimizers\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 857\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m_propose_text_optimizers\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 858\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mtext_optimizer\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtext_optimizers\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 859\u001b[0;31m \u001b[0mtext_optimizer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpropose\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 860\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 861\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m_get_trainable_text_params\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/adalflow/optim/text_grad/tgd_optimizer.py\u001b[0m in \u001b[0;36mpropose\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 323\u001b[0m }\n\u001b[1;32m 324\u001b[0m \u001b[0;31m# turn off cache\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 325\u001b[0;31m response = self.llm_optimizer.call(\n\u001b[0m\u001b[1;32m 326\u001b[0m \u001b[0mprompt_kwargs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mprompt_kwargs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0muse_cache\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mnot\u001b[0m \u001b[0mno_cache\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 327\u001b[0m )\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/adalflow/core/generator.py\u001b[0m in \u001b[0;36mcall\u001b[0;34m(self, prompt_kwargs, model_kwargs, use_cache, id)\u001b[0m\n\u001b[1;32m 771\u001b[0m \u001b[0;31m# call the model client\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 772\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 773\u001b[0;31m \u001b[0mcompletion\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 774\u001b[0m \u001b[0muse_cache\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0muse_cache\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0muse_cache\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_use_cache\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 775\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/adalflow/core/generator.py\u001b[0m in \u001b[0;36m_model_client_call\u001b[0;34m(self, api_kwargs, use_cache)\u001b[0m\n\u001b[1;32m 345\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 346\u001b[0m \u001b[0mcached_completion\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_check_cache\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mindex_content\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 347\u001b[0;31m \u001b[0;32mif\u001b[0m \u001b[0mcached_completion\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 348\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mcached_completion\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 349\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/backoff/_sync.py\u001b[0m in \u001b[0;36mretry\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m 103\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 104\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 105\u001b[0;31m \u001b[0mret\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtarget\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 106\u001b[0m \u001b[0;32mexcept\u001b[0m \u001b[0mexception\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 107\u001b[0m \u001b[0mmax_tries_exceeded\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mtries\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0mmax_tries_value\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/adalflow/components/model_client/openai_client.py\u001b[0m in \u001b[0;36mcall\u001b[0;34m(self, api_kwargs, model_type)\u001b[0m\n\u001b[1;32m 285\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mchat_completion_parser\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mhandle_streaming_response\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 286\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msync_client\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mchat\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcompletions\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcreate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0mapi_kwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 287\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msync_client\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mchat\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcompletions\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcreate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0mapi_kwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 288\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 289\u001b[0m \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf\"model_type {model_type} is not supported\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/openai/_utils/_utils.py\u001b[0m in \u001b[0;36mwrapper\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m 273\u001b[0m \u001b[0mmsg\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34mf\"Missing required argument: {quote(missing[0])}\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 274\u001b[0m \u001b[0;32mraise\u001b[0m \u001b[0mTypeError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmsg\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 275\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 276\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 277\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mwrapper\u001b[0m \u001b[0;31m# type: ignore\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/openai/resources/chat/completions.py\u001b[0m in \u001b[0;36mcreate\u001b[0;34m(self, messages, model, audio, frequency_penalty, function_call, functions, logit_bias, logprobs, max_completion_tokens, max_tokens, metadata, modalities, n, parallel_tool_calls, prediction, presence_penalty, response_format, seed, service_tier, stop, store, stream, stream_options, temperature, tool_choice, tools, top_logprobs, top_p, user, extra_headers, extra_query, extra_body, timeout)\u001b[0m\n\u001b[1;32m 827\u001b[0m ) -> ChatCompletion | Stream[ChatCompletionChunk]:\n\u001b[1;32m 828\u001b[0m \u001b[0mvalidate_response_format\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mresponse_format\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 829\u001b[0;31m return self._post(\n\u001b[0m\u001b[1;32m 830\u001b[0m \u001b[0;34m\"/chat/completions\"\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 831\u001b[0m body=maybe_transform(\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/openai/_base_client.py\u001b[0m in \u001b[0;36mpost\u001b[0;34m(self, path, cast_to, body, options, files, stream, stream_cls)\u001b[0m\n\u001b[1;32m 1276\u001b[0m \u001b[0mmethod\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m\"post\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0murl\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mpath\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mjson_data\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mbody\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfiles\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mto_httpx_files\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfiles\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0moptions\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1277\u001b[0m )\n\u001b[0;32m-> 1278\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mcast\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mResponseT\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrequest\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcast_to\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mopts\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstream\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mstream\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstream_cls\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mstream_cls\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1279\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1280\u001b[0m def patch(\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/openai/_base_client.py\u001b[0m in \u001b[0;36mrequest\u001b[0;34m(self, cast_to, options, remaining_retries, stream, stream_cls)\u001b[0m\n\u001b[1;32m 953\u001b[0m \u001b[0mretries_taken\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 954\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 955\u001b[0;31m return self._request(\n\u001b[0m\u001b[1;32m 956\u001b[0m \u001b[0mcast_to\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcast_to\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 957\u001b[0m \u001b[0moptions\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0moptions\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/openai/_base_client.py\u001b[0m in \u001b[0;36m_request\u001b[0;34m(self, cast_to, options, retries_taken, stream, stream_cls)\u001b[0m\n\u001b[1;32m 989\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 990\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 991\u001b[0;31m response = self._client.send(\n\u001b[0m\u001b[1;32m 992\u001b[0m \u001b[0mrequest\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 993\u001b[0m \u001b[0mstream\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mstream\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_should_stream_response_body\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrequest\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mrequest\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/httpx/_client.py\u001b[0m in \u001b[0;36msend\u001b[0;34m(self, request, stream, auth, follow_redirects)\u001b[0m\n\u001b[1;32m 899\u001b[0m \u001b[0mauth\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_build_request_auth\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrequest\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mauth\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 900\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 901\u001b[0;31m response = self._send_handling_auth(\n\u001b[0m\u001b[1;32m 902\u001b[0m \u001b[0mrequest\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 903\u001b[0m \u001b[0mauth\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mauth\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/httpx/_client.py\u001b[0m in \u001b[0;36m_send_handling_auth\u001b[0;34m(self, request, auth, follow_redirects, history)\u001b[0m\n\u001b[1;32m 927\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 928\u001b[0m \u001b[0;32mwhile\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 929\u001b[0;31m response = self._send_handling_redirects(\n\u001b[0m\u001b[1;32m 930\u001b[0m \u001b[0mrequest\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 931\u001b[0m \u001b[0mfollow_redirects\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mfollow_redirects\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/httpx/_client.py\u001b[0m in \u001b[0;36m_send_handling_redirects\u001b[0;34m(self, request, follow_redirects, history)\u001b[0m\n\u001b[1;32m 964\u001b[0m \u001b[0mhook\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrequest\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 965\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 966\u001b[0;31m \u001b[0mresponse\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_send_single_request\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrequest\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 967\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 968\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mhook\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_event_hooks\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"response\"\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/httpx/_client.py\u001b[0m in \u001b[0;36m_send_single_request\u001b[0;34m(self, request)\u001b[0m\n\u001b[1;32m 1000\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1001\u001b[0m \u001b[0;32mwith\u001b[0m \u001b[0mrequest_context\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrequest\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mrequest\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1002\u001b[0;31m \u001b[0mresponse\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtransport\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhandle_request\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrequest\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1003\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1004\u001b[0m \u001b[0;32massert\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mresponse\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstream\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mSyncByteStream\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/httpx/_transports/default.py\u001b[0m in \u001b[0;36mhandle_request\u001b[0;34m(self, request)\u001b[0m\n\u001b[1;32m 216\u001b[0m )\n\u001b[1;32m 217\u001b[0m \u001b[0;32mwith\u001b[0m \u001b[0mmap_httpcore_exceptions\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 218\u001b[0;31m \u001b[0mresp\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_pool\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhandle_request\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mreq\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 219\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 220\u001b[0m \u001b[0;32massert\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mresp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstream\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtyping\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mIterable\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/httpcore/_sync/connection_pool.py\u001b[0m in \u001b[0;36mhandle_request\u001b[0;34m(self, request)\u001b[0m\n\u001b[1;32m 260\u001b[0m \u001b[0;32mwith\u001b[0m \u001b[0mShieldCancellation\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 261\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mresponse_closed\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mstatus\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 262\u001b[0;31m \u001b[0;32mraise\u001b[0m \u001b[0mexc\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 263\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 264\u001b[0m \u001b[0;32mbreak\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/httpcore/_sync/connection_pool.py\u001b[0m in \u001b[0;36mhandle_request\u001b[0;34m(self, request)\u001b[0m\n\u001b[1;32m 243\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 244\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 245\u001b[0;31m \u001b[0mresponse\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mconnection\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhandle_request\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrequest\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 246\u001b[0m \u001b[0;32mexcept\u001b[0m \u001b[0mConnectionNotAvailable\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 247\u001b[0m \u001b[0;31m# The ConnectionNotAvailable exception is a special case, that\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/httpcore/_sync/connection.py\u001b[0m in \u001b[0;36mhandle_request\u001b[0;34m(self, request)\u001b[0m\n\u001b[1;32m 94\u001b[0m \u001b[0;32mraise\u001b[0m \u001b[0mConnectionNotAvailable\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 95\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 96\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_connection\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhandle_request\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrequest\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 97\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 98\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m_connect\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mrequest\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mRequest\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m->\u001b[0m \u001b[0mNetworkStream\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/httpcore/_sync/http11.py\u001b[0m in \u001b[0;36mhandle_request\u001b[0;34m(self, request)\u001b[0m\n\u001b[1;32m 119\u001b[0m \u001b[0;32mwith\u001b[0m \u001b[0mTrace\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"response_closed\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlogger\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mrequest\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mtrace\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 120\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_response_closed\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 121\u001b[0;31m \u001b[0;32mraise\u001b[0m \u001b[0mexc\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 122\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 123\u001b[0m \u001b[0;31m# Sending the request...\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/httpcore/_sync/http11.py\u001b[0m in \u001b[0;36mhandle_request\u001b[0;34m(self, request)\u001b[0m\n\u001b[1;32m 97\u001b[0m \u001b[0mreason_phrase\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 98\u001b[0m \u001b[0mheaders\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 99\u001b[0;31m ) = self._receive_response_headers(**kwargs)\n\u001b[0m\u001b[1;32m 100\u001b[0m trace.return_value = (\n\u001b[1;32m 101\u001b[0m \u001b[0mhttp_version\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/httpcore/_sync/http11.py\u001b[0m in \u001b[0;36m_receive_response_headers\u001b[0;34m(self, request)\u001b[0m\n\u001b[1;32m 162\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 163\u001b[0m \u001b[0;32mwhile\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 164\u001b[0;31m \u001b[0mevent\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_receive_event\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtimeout\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtimeout\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 165\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mevent\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mh11\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mResponse\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 166\u001b[0m \u001b[0;32mbreak\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/httpcore/_sync/http11.py\u001b[0m in \u001b[0;36m_receive_event\u001b[0;34m(self, timeout)\u001b[0m\n\u001b[1;32m 198\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 199\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mevent\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0mh11\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mNEED_DATA\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 200\u001b[0;31m data = self._network_stream.read(\n\u001b[0m\u001b[1;32m 201\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mREAD_NUM_BYTES\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtimeout\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtimeout\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 202\u001b[0m )\n",
+ "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/httpcore/_backends/sync.py\u001b[0m in \u001b[0;36mread\u001b[0;34m(self, max_bytes, timeout)\u001b[0m\n\u001b[1;32m 26\u001b[0m \u001b[0;32mwith\u001b[0m \u001b[0mmap_exceptions\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mexc_map\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 27\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_sock\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msettimeout\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtimeout\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 28\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_sock\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrecv\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmax_bytes\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 29\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 30\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mwrite\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbuffer\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mbytes\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtimeout\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mtyping\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mOptional\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mfloat\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m->\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/lib/python3.10/ssl.py\u001b[0m in \u001b[0;36mrecv\u001b[0;34m(self, buflen, flags)\u001b[0m\n\u001b[1;32m 1286\u001b[0m \u001b[0;34m\"non-zero flags not allowed in calls to recv() on %s\"\u001b[0m \u001b[0;34m%\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1287\u001b[0m self.__class__)\n\u001b[0;32m-> 1288\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbuflen\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1289\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1290\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0msuper\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrecv\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbuflen\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mflags\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m/usr/lib/python3.10/ssl.py\u001b[0m in \u001b[0;36mread\u001b[0;34m(self, len, buffer)\u001b[0m\n\u001b[1;32m 1159\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_sslobj\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbuffer\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1160\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1161\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_sslobj\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1162\u001b[0m \u001b[0;32mexcept\u001b[0m \u001b[0mSSLError\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1163\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0mSSL_ERROR_EOF\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msuppress_ragged_eofs\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
+ ]
+ }
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "source": [
+ "# Issues and feedback\n",
+ "\n",
+ "If you encounter any issues, please report them here: [GitHub Issues](https://github.com/SylphAI-Inc/LightRAG/issues).\n",
+ "\n",
+ "For feedback, you can use either the [GitHub discussions](https://github.com/SylphAI-Inc/LightRAG/discussions) or [Discord](https://discord.gg/ezzszrRZvT)."
+ ],
+ "metadata": {
+ "id": "AmkbyxmuruUu"
+ }
+ }
+ ]
}
diff --git a/notebooks/tutorials/adalflow_component.ipynb b/notebooks/tutorials/adalflow_component.ipynb
index 8523a629..66050d07 100644
--- a/notebooks/tutorials/adalflow_component.ipynb
+++ b/notebooks/tutorials/adalflow_component.ipynb
@@ -59,6 +59,17 @@
"clear_output()"
]
},
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "!pip uninstall httpx anyio -y\n",
+ "!pip install \"anyio>=3.1.0,<4.0\"\n",
+ "!pip install httpx==0.24.1"
+ ]
+ },
{
"cell_type": "code",
"execution_count": 5,
diff --git a/notebooks/tutorials/adalflow_dataclasses.ipynb b/notebooks/tutorials/adalflow_dataclasses.ipynb
index 7ae08f63..db35d95e 100644
--- a/notebooks/tutorials/adalflow_dataclasses.ipynb
+++ b/notebooks/tutorials/adalflow_dataclasses.ipynb
@@ -82,6 +82,17 @@
"clear_output()"
]
},
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "!pip uninstall httpx anyio -y\n",
+ "!pip install \"anyio>=3.1.0,<4.0\"\n",
+ "!pip install httpx==0.24.1"
+ ]
+ },
{
"cell_type": "markdown",
"metadata": {
diff --git a/notebooks/tutorials/adalflow_function_calls.ipynb b/notebooks/tutorials/adalflow_function_calls.ipynb
index 6fba3594..ee53cf07 100644
--- a/notebooks/tutorials/adalflow_function_calls.ipynb
+++ b/notebooks/tutorials/adalflow_function_calls.ipynb
@@ -1,21 +1,10 @@
{
- "nbformat": 4,
- "nbformat_minor": 0,
- "metadata": {
- "colab": {
- "provenance": []
- },
- "kernelspec": {
- "name": "python3",
- "display_name": "Python 3"
- },
- "language_info": {
- "name": "python"
- }
- },
"cells": [
{
"cell_type": "markdown",
+ "metadata": {
+ "id": "lLGpv1fLLIjF"
+ },
"source": [
"# Function calls\n",
"\n",
@@ -30,10 +19,7 @@
"- Function call in action\n",
"\n",
"It follows the tutorial here: https://adalflow.sylph.ai/tutorials/tool_helper.html#"
- ],
- "metadata": {
- "id": "lLGpv1fLLIjF"
- }
+ ]
},
{
"cell_type": "code",
@@ -52,20 +38,18 @@
},
{
"cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
"source": [
- "import os\n",
- "from getpass import getpass\n",
- "\n",
- "# Prompt user to enter their API keys securely\n",
- "openai_api_key = getpass(\"Please enter your OpenAI API key: \")\n",
- "groq_api_key = getpass(\"Please enter your GROQ API key: \")\n",
- "\n",
- "# Set environment variables\n",
- "os.environ[\"OPENAI_API_KEY\"] = openai_api_key\n",
- "os.environ[\"GROQ_API_KEY\"] = groq_api_key\n",
- "\n",
- "print(\"API keys have been set.\")"
- ],
+ "!pip uninstall httpx anyio -y\n",
+ "!pip install \"anyio>=3.1.0,<4.0\"\n",
+ "!pip install httpx==0.24.1"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
@@ -73,21 +57,39 @@
"id": "-4c_AGBt3PlR",
"outputId": "21a26437-9f95-4478-84e9-ba4369956b6f"
},
- "execution_count": 2,
"outputs": [
{
- "output_type": "stream",
"name": "stdout",
+ "output_type": "stream",
"text": [
"Please enter your OpenAI API key: ยทยทยทยทยทยทยทยทยทยท\n",
"Please enter your GROQ API key: ยทยทยทยทยทยทยทยทยทยท\n",
"API keys have been set.\n"
]
}
+ ],
+ "source": [
+ "import os\n",
+ "from getpass import getpass\n",
+ "\n",
+ "# Prompt user to enter their API keys securely\n",
+ "openai_api_key = getpass(\"Please enter your OpenAI API key: \")\n",
+ "groq_api_key = getpass(\"Please enter your GROQ API key: \")\n",
+ "\n",
+ "# Set environment variables\n",
+ "os.environ[\"OPENAI_API_KEY\"] = openai_api_key\n",
+ "os.environ[\"GROQ_API_KEY\"] = groq_api_key\n",
+ "\n",
+ "print(\"API keys have been set.\")"
]
},
{
"cell_type": "code",
+ "execution_count": 4,
+ "metadata": {
+ "id": "GMKuuP7xR9Nt"
+ },
+ "outputs": [],
"source": [
"from dataclasses import dataclass\n",
"from typing import List\n",
@@ -136,32 +138,20 @@
"\n",
"def add_points(p1: Point, p2: Point) -> Point:\n",
" return Point(p1.x + p2.x, p1.y + p2.y)"
- ],
- "metadata": {
- "id": "GMKuuP7xR9Nt"
- },
- "execution_count": 4,
- "outputs": []
+ ]
},
{
"cell_type": "markdown",
- "source": [
- "##ย Function Tool"
- ],
"metadata": {
"id": "jCA7HMjtT16P"
- }
+ },
+ "source": [
+ "##ย Function Tool"
+ ]
},
{
"cell_type": "code",
- "source": [
- "from adalflow.core.func_tool import FunctionTool\n",
- "\n",
- "functions = [multiply, add, divide, search, numpy_sum, add_points]\n",
- "tools = [FunctionTool(fn=fn) for fn in functions]\n",
- "for tool in tools:\n",
- " print(tool)"
- ],
+ "execution_count": 5,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
@@ -169,11 +159,10 @@
"id": "fgOEoLoDSBqh",
"outputId": "7e636e2c-9a5d-44f1-f0fe-fe8a6bea474d"
},
- "execution_count": 5,
"outputs": [
{
- "output_type": "stream",
"name": "stdout",
+ "output_type": "stream",
"text": [
"FunctionTool(fn: , async: False, definition: FunctionDefinition(func_name='multiply', func_desc='multiply(a: int, b: int) -> int\\nMultiply two numbers.', func_parameters={'type': 'object', 'properties': {'a': {'type': 'int'}, 'b': {'type': 'int'}}, 'required': ['a', 'b']}))\n",
"FunctionTool(fn: , async: False, definition: FunctionDefinition(func_name='add', func_desc='add(a: int, b: int) -> int\\nAdd two numbers.', func_parameters={'type': 'object', 'properties': {'a': {'type': 'int'}, 'b': {'type': 'int'}}, 'required': ['a', 'b']}))\n",
@@ -183,13 +172,19 @@
"FunctionTool(fn: , async: False, definition: FunctionDefinition(func_name='add_points', func_desc='add_points(p1: __main__.Point, p2: __main__.Point) -> __main__.Point\\nNone', func_parameters={'type': 'object', 'properties': {'p1': {'type': \"{'type': 'Point', 'properties': {'x': {'type': 'int'}, 'y': {'type': 'int'}}, 'required': ['x', 'y']}\"}, 'p2': {'type': \"{'type': 'Point', 'properties': {'x': {'type': 'int'}, 'y': {'type': 'int'}}, 'required': ['x', 'y']}\"}}, 'required': ['p1', 'p2']}))\n"
]
}
+ ],
+ "source": [
+ "from adalflow.core.func_tool import FunctionTool\n",
+ "\n",
+ "functions = [multiply, add, divide, search, numpy_sum, add_points]\n",
+ "tools = [FunctionTool(fn=fn) for fn in functions]\n",
+ "for tool in tools:\n",
+ " print(tool)"
]
},
{
"cell_type": "code",
- "source": [
- "print(tools[-2].definition.to_dict())"
- ],
+ "execution_count": 6,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
@@ -197,50 +192,47 @@
"id": "CYJaHFhGSEzH",
"outputId": "9ab36c6c-7509-4e7f-ce85-11dae889c8c2"
},
- "execution_count": 6,
"outputs": [
{
- "output_type": "stream",
"name": "stdout",
+ "output_type": "stream",
"text": [
"{'func_name': 'numpy_sum', 'func_desc': 'numpy_sum(arr: numpy.ndarray) -> float\\nSum the elements of an array.', 'func_parameters': {'type': 'object', 'properties': {'arr': {'type': 'ndarray'}}, 'required': ['arr']}}\n"
]
}
+ ],
+ "source": [
+ "print(tools[-2].definition.to_dict())"
]
},
{
"cell_type": "code",
- "source": [
- "context_map = {tool.definition.func_name: tool for tool in tools}"
- ],
+ "execution_count": 7,
"metadata": {
"id": "_O4bQgXrSKb6"
},
- "execution_count": 7,
- "outputs": []
+ "outputs": [],
+ "source": [
+ "context_map = {tool.definition.func_name: tool for tool in tools}"
+ ]
},
{
"cell_type": "code",
+ "execution_count": 8,
+ "metadata": {
+ "id": "-RgWWMdISL1u"
+ },
+ "outputs": [],
"source": [
"function_name = \"add\"\n",
"function_to_call = context_map[function_name]\n",
"function_args = {\"a\": 1, \"b\": 2}\n",
"function_response = function_to_call.call(**function_args)"
- ],
- "metadata": {
- "id": "-RgWWMdISL1u"
- },
- "execution_count": 8,
- "outputs": []
+ ]
},
{
"cell_type": "code",
- "source": [
- "from adalflow.core.tool_manager import ToolManager\n",
- "\n",
- "tool_manager = ToolManager(tools=functions)\n",
- "print(tool_manager)"
- ],
+ "execution_count": 9,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
@@ -248,34 +240,34 @@
"id": "6CT7Tez1SOai",
"outputId": "e486d882-9179-4db3-f077-6adfc9fc6579"
},
- "execution_count": 9,
"outputs": [
{
- "output_type": "stream",
"name": "stdout",
+ "output_type": "stream",
"text": [
"ToolManager(Tools: [FunctionTool(fn: , async: False, definition: FunctionDefinition(func_name='multiply', func_desc='multiply(a: int, b: int) -> int\\nMultiply two numbers.', func_parameters={'type': 'object', 'properties': {'a': {'type': 'int'}, 'b': {'type': 'int'}}, 'required': ['a', 'b']})), FunctionTool(fn: , async: False, definition: FunctionDefinition(func_name='add', func_desc='add(a: int, b: int) -> int\\nAdd two numbers.', func_parameters={'type': 'object', 'properties': {'a': {'type': 'int'}, 'b': {'type': 'int'}}, 'required': ['a', 'b']})), FunctionTool(fn: , async: True, definition: FunctionDefinition(func_name='divide', func_desc='divide(a: float, b: float) -> float\\nDivide two numbers.', func_parameters={'type': 'object', 'properties': {'a': {'type': 'float'}, 'b': {'type': 'float'}}, 'required': ['a', 'b']})), FunctionTool(fn: , async: True, definition: FunctionDefinition(func_name='search', func_desc='search(query: str) -> List[str]\\nSearch for query and return a list of results.', func_parameters={'type': 'object', 'properties': {'query': {'type': 'str'}}, 'required': ['query']})), FunctionTool(fn: , async: False, definition: FunctionDefinition(func_name='numpy_sum', func_desc='numpy_sum(arr: numpy.ndarray) -> float\\nSum the elements of an array.', func_parameters={'type': 'object', 'properties': {'arr': {'type': 'ndarray'}}, 'required': ['arr']})), FunctionTool(fn: , async: False, definition: FunctionDefinition(func_name='add_points', func_desc='add_points(p1: __main__.Point, p2: __main__.Point) -> __main__.Point\\nNone', func_parameters={'type': 'object', 'properties': {'p1': {'type': \"{'type': 'Point', 'properties': {'x': {'type': 'int'}, 'y': {'type': 'int'}}, 'required': ['x', 'y']}\"}, 'p2': {'type': \"{'type': 'Point', 'properties': {'x': {'type': 'int'}, 'y': {'type': 'int'}}, 'required': ['x', 'y']}\"}}, 'required': ['p1', 'p2']}))], Additional Context: {})\n"
]
}
+ ],
+ "source": [
+ "from adalflow.core.tool_manager import ToolManager\n",
+ "\n",
+ "tool_manager = ToolManager(tools=functions)\n",
+ "print(tool_manager)"
]
},
{
"cell_type": "markdown",
- "source": [
- "## ToolManager"
- ],
"metadata": {
"id": "jzFqNnN_T-cu"
- }
+ },
+ "source": [
+ "## ToolManager"
+ ]
},
{
"cell_type": "code",
- "source": [
- "from adalflow.core.tool_manager import ToolManager\n",
- "\n",
- "tool_manager = ToolManager(tools=functions)\n",
- "print(tool_manager)"
- ],
+ "execution_count": 10,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
@@ -283,28 +275,38 @@
"id": "JX7MibWiUF3U",
"outputId": "20707186-5ec3-49a4-d553-c3160c3daa84"
},
- "execution_count": 10,
"outputs": [
{
- "output_type": "stream",
"name": "stdout",
+ "output_type": "stream",
"text": [
"ToolManager(Tools: [FunctionTool(fn: , async: False, definition: FunctionDefinition(func_name='multiply', func_desc='multiply(a: int, b: int) -> int\\nMultiply two numbers.', func_parameters={'type': 'object', 'properties': {'a': {'type': 'int'}, 'b': {'type': 'int'}}, 'required': ['a', 'b']})), FunctionTool(fn: , async: False, definition: FunctionDefinition(func_name='add', func_desc='add(a: int, b: int) -> int\\nAdd two numbers.', func_parameters={'type': 'object', 'properties': {'a': {'type': 'int'}, 'b': {'type': 'int'}}, 'required': ['a', 'b']})), FunctionTool(fn: , async: True, definition: FunctionDefinition(func_name='divide', func_desc='divide(a: float, b: float) -> float\\nDivide two numbers.', func_parameters={'type': 'object', 'properties': {'a': {'type': 'float'}, 'b': {'type': 'float'}}, 'required': ['a', 'b']})), FunctionTool(fn: , async: True, definition: FunctionDefinition(func_name='search', func_desc='search(query: str) -> List[str]\\nSearch for query and return a list of results.', func_parameters={'type': 'object', 'properties': {'query': {'type': 'str'}}, 'required': ['query']})), FunctionTool(fn: , async: False, definition: FunctionDefinition(func_name='numpy_sum', func_desc='numpy_sum(arr: numpy.ndarray) -> float\\nSum the elements of an array.', func_parameters={'type': 'object', 'properties': {'arr': {'type': 'ndarray'}}, 'required': ['arr']})), FunctionTool(fn: , async: False, definition: FunctionDefinition(func_name='add_points', func_desc='add_points(p1: __main__.Point, p2: __main__.Point) -> __main__.Point\\nNone', func_parameters={'type': 'object', 'properties': {'p1': {'type': \"{'type': 'Point', 'properties': {'x': {'type': 'int'}, 'y': {'type': 'int'}}, 'required': ['x', 'y']}\"}, 'p2': {'type': \"{'type': 'Point', 'properties': {'x': {'type': 'int'}, 'y': {'type': 'int'}}, 'required': ['x', 'y']}\"}}, 'required': ['p1', 'p2']}))], Additional Context: {})\n"
]
}
+ ],
+ "source": [
+ "from adalflow.core.tool_manager import ToolManager\n",
+ "\n",
+ "tool_manager = ToolManager(tools=functions)\n",
+ "print(tool_manager)"
]
},
{
"cell_type": "markdown",
- "source": [
- "## Function Call end-to-end"
- ],
"metadata": {
"id": "9Bw2fs--UKX7"
- }
+ },
+ "source": [
+ "## Function Call end-to-end"
+ ]
},
{
"cell_type": "code",
+ "execution_count": 11,
+ "metadata": {
+ "id": "TywPQMIVUOqh"
+ },
+ "outputs": [],
"source": [
"template = r\"\"\"You have these tools available:\n",
"{% if tools %}\n",
@@ -323,24 +325,11 @@
"User: {{input_str}}\n",
"You:\n",
"\"\"\""
- ],
- "metadata": {
- "id": "TywPQMIVUOqh"
- },
- "execution_count": 11,
- "outputs": []
+ ]
},
{
"cell_type": "code",
- "source": [
- "from adalflow.core.prompt_builder import Prompt\n",
- "\n",
- "prompt = Prompt(template=template)\n",
- "small_tool_manager = ToolManager(tools=tools[:2])\n",
- "\n",
- "renered_prompt = prompt(tools=small_tool_manager.yaml_definitions)\n",
- "print(renered_prompt)"
- ],
+ "execution_count": 12,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
@@ -348,11 +337,10 @@
"id": "-vMajeXoUQ5A",
"outputId": "ca68601b-e9c8-41c3-a6fa-777f225e68e3"
},
- "execution_count": 12,
"outputs": [
{
- "output_type": "stream",
"name": "stdout",
+ "output_type": "stream",
"text": [
"You have these tools available:\n",
"\n",
@@ -398,19 +386,20 @@
"\n"
]
}
- ]
- },
- {
- "cell_type": "code",
+ ],
"source": [
- "from adalflow.core.types import Function\n",
+ "from adalflow.core.prompt_builder import Prompt\n",
"\n",
- "output_data_class = Function\n",
- "output_format_str = output_data_class.to_json_signature(exclude=[\"thought\", \"args\"])\n",
+ "prompt = Prompt(template=template)\n",
+ "small_tool_manager = ToolManager(tools=tools[:2])\n",
"\n",
- "renered_prompt = prompt(output_format_str=output_format_str)\n",
+ "renered_prompt = prompt(tools=small_tool_manager.yaml_definitions)\n",
"print(renered_prompt)"
- ],
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 13,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
@@ -418,11 +407,10 @@
"id": "V9-90IFRUUNT",
"outputId": "ed2f829e-c656-43c6-a454-8a7c32d5dafe"
},
- "execution_count": 13,
"outputs": [
{
- "output_type": "stream",
"name": "stdout",
+ "output_type": "stream",
"text": [
"You have these tools available:\n",
"\n",
@@ -437,17 +425,20 @@
"\n"
]
}
+ ],
+ "source": [
+ "from adalflow.core.types import Function\n",
+ "\n",
+ "output_data_class = Function\n",
+ "output_format_str = output_data_class.to_json_signature(exclude=[\"thought\", \"args\"])\n",
+ "\n",
+ "renered_prompt = prompt(output_format_str=output_format_str)\n",
+ "print(renered_prompt)"
]
},
{
"cell_type": "code",
- "source": [
- "from adalflow.core.types import FunctionExpression\n",
- "\n",
- "output_data_class = FunctionExpression\n",
- "output_format_str = output_data_class.to_json_signature(exclude=[\"thought\"])\n",
- "print(prompt(output_format_str=output_format_str))"
- ],
+ "execution_count": 14,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
@@ -455,11 +446,10 @@
"id": "p3kPMhWaUYT1",
"outputId": "a3de7117-c3eb-404e-e2e7-8a5187b32f6b"
},
- "execution_count": 14,
"outputs": [
{
- "output_type": "stream",
"name": "stdout",
+ "output_type": "stream",
"text": [
"You have these tools available:\n",
"\n",
@@ -473,17 +463,18 @@
"\n"
]
}
+ ],
+ "source": [
+ "from adalflow.core.types import FunctionExpression\n",
+ "\n",
+ "output_data_class = FunctionExpression\n",
+ "output_format_str = output_data_class.to_json_signature(exclude=[\"thought\"])\n",
+ "print(prompt(output_format_str=output_format_str))"
]
},
{
"cell_type": "code",
- "source": [
- "from adalflow.components.output_parsers import JsonOutputParser\n",
- "\n",
- "func_parser = JsonOutputParser(data_class=Function, exclude_fields=[\"thought\", \"args\"])\n",
- "instructions = func_parser.format_instructions()\n",
- "print(instructions)"
- ],
+ "execution_count": 17,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
@@ -491,11 +482,10 @@
"id": "MvGyoUmMUatR",
"outputId": "e819866b-f6e3-4c88-f9f1-22d725a28865"
},
- "execution_count": 17,
"outputs": [
{
- "output_type": "stream",
"name": "stdout",
+ "output_type": "stream",
"text": [
"Your output should be formatted as a standard JSON instance with the following schema:\n",
"```\n",
@@ -510,19 +500,31 @@
"-Follow the JSON formatting conventions.\n"
]
}
+ ],
+ "source": [
+ "from adalflow.components.output_parsers import JsonOutputParser\n",
+ "\n",
+ "func_parser = JsonOutputParser(data_class=Function, exclude_fields=[\"thought\", \"args\"])\n",
+ "instructions = func_parser.format_instructions()\n",
+ "print(instructions)"
]
},
{
"cell_type": "markdown",
- "source": [
- "## Function Output Format"
- ],
"metadata": {
"id": "9W7DiGcpUme5"
- }
+ },
+ "source": [
+ "## Function Output Format"
+ ]
},
{
"cell_type": "code",
+ "execution_count": 20,
+ "metadata": {
+ "id": "z5tNhoruUp6o"
+ },
+ "outputs": [],
"source": [
"from adalflow.core.generator import Generator\n",
"from adalflow.core.types import ModelClientType\n",
@@ -539,42 +541,11 @@
" prompt_kwargs=prompt_kwargs,\n",
" output_processors=func_parser,\n",
")"
- ],
- "metadata": {
- "id": "z5tNhoruUp6o"
- },
- "execution_count": 20,
- "outputs": []
+ ]
},
{
"cell_type": "code",
- "source": [
- "queries = [\n",
- " \"add 2 and 3\",\n",
- " \"search for something\",\n",
- " \"add points (1, 2) and (3, 4)\",\n",
- " \"sum numpy array with arr = np.array([[1, 2], [3, 4]])\",\n",
- " \"multiply 2 with local variable x\",\n",
- " \"divide 2 by 3\",\n",
- " \"Add 5 to variable y\",\n",
- "]\n",
- "\n",
- "for idx, query in enumerate(queries):\n",
- " prompt_kwargs = {\"input_str\": query}\n",
- " print(f\"\\n{idx} Query: {query}\")\n",
- " print(f\"{'-'*50}\")\n",
- " try:\n",
- " result = generator(prompt_kwargs=prompt_kwargs)\n",
- " # print(f\"LLM raw output: {result.raw_response}\")\n",
- " func = Function.from_dict(result.data)\n",
- " print(f\"Function: {func}\")\n",
- " func_output = tool_manager.execute_func(func)\n",
- " print(f\"Function output: {func_output}\")\n",
- " except Exception as e:\n",
- " print(\n",
- " f\"Failed to execute the function for query: {query}, func: {result.data}, error: {e}\"\n",
- " )"
- ],
+ "execution_count": 21,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
@@ -582,11 +553,10 @@
"id": "9DCukn1SUs_x",
"outputId": "dcfd952c-0699-4d79-ee6d-a59373e3c75d"
},
- "execution_count": 21,
"outputs": [
{
- "output_type": "stream",
"name": "stdout",
+ "output_type": "stream",
"text": [
"\n",
"0 Query: add 2 and 3\n",
@@ -604,15 +574,15 @@
]
},
{
- "output_type": "stream",
"name": "stderr",
+ "output_type": "stream",
"text": [
"ERROR:adalflow.core.func_tool:Error at calling : 'dict' object has no attribute 'x'\n"
]
},
{
- "output_type": "stream",
"name": "stdout",
+ "output_type": "stream",
"text": [
"Function: Function(thought=None, name='add_points', args=[], kwargs={'p1': {'x': 1, 'y': 2}, 'p2': {'x': 3, 'y': 4}})\n",
"Function output: FunctionOutput(name='add_points', input=Function(thought=None, name='add_points', args=(), kwargs={'p1': {'x': 1, 'y': 2}, 'p2': {'x': 3, 'y': 4}}), parsed_input=None, output=None, error=\"'dict' object has no attribute 'x'\")\n",
@@ -638,62 +608,94 @@
]
},
{
- "output_type": "stream",
"name": "stderr",
+ "output_type": "stream",
"text": [
"ERROR:adalflow.core.func_tool:Error at calling : unsupported operand type(s) for +: 'int' and 'str'\n"
]
},
{
- "output_type": "stream",
"name": "stdout",
+ "output_type": "stream",
"text": [
"Function output: FunctionOutput(name='add', input=Function(thought=None, name='add', args=(), kwargs={'a': 5, 'b': 'y'}), parsed_input=None, output=None, error=\"unsupported operand type(s) for +: 'int' and 'str'\")\n"
]
}
+ ],
+ "source": [
+ "queries = [\n",
+ " \"add 2 and 3\",\n",
+ " \"search for something\",\n",
+ " \"add points (1, 2) and (3, 4)\",\n",
+ " \"sum numpy array with arr = np.array([[1, 2], [3, 4]])\",\n",
+ " \"multiply 2 with local variable x\",\n",
+ " \"divide 2 by 3\",\n",
+ " \"Add 5 to variable y\",\n",
+ "]\n",
+ "\n",
+ "for idx, query in enumerate(queries):\n",
+ " prompt_kwargs = {\"input_str\": query}\n",
+ " print(f\"\\n{idx} Query: {query}\")\n",
+ " print(f\"{'-'*50}\")\n",
+ " try:\n",
+ " result = generator(prompt_kwargs=prompt_kwargs)\n",
+ " # print(f\"LLM raw output: {result.raw_response}\")\n",
+ " func = Function.from_dict(result.data)\n",
+ " print(f\"Function: {func}\")\n",
+ " func_output = tool_manager.execute_func(func)\n",
+ " print(f\"Function output: {func_output}\")\n",
+ " except Exception as e:\n",
+ " print(\n",
+ " f\"Failed to execute the function for query: {query}, func: {result.data}, error: {e}\"\n",
+ " )"
]
},
{
"cell_type": "markdown",
- "source": [
- "## FunctionExpression Output Format"
- ],
"metadata": {
"id": "O-sBTPATUwsD"
- }
+ },
+ "source": [
+ "## FunctionExpression Output Format"
+ ]
},
{
"cell_type": "code",
+ "execution_count": 22,
+ "metadata": {
+ "id": "TVRZ44N1UyWg"
+ },
+ "outputs": [],
"source": [
"tool_manager = ToolManager(\n",
" tools=functions,\n",
" additional_context={\"x\": x, \"y\": 0, \"np.array\": np.array, \"np\": np},\n",
")\n",
"func_parser = JsonOutputParser(data_class=FunctionExpression)"
- ],
- "metadata": {
- "id": "TVRZ44N1UyWg"
- },
- "execution_count": 22,
- "outputs": []
+ ]
},
{
"cell_type": "code",
+ "execution_count": 23,
+ "metadata": {
+ "id": "9h47p4XpU2BC"
+ },
+ "outputs": [],
"source": [
"context = r\"\"\"\n",
"Your function expression also have access to these context:\n",
"{{context_str}}\n",
"\n",
"\"\"\""
- ],
- "metadata": {
- "id": "9h47p4XpU2BC"
- },
- "execution_count": 23,
- "outputs": []
+ ]
},
{
"cell_type": "code",
+ "execution_count": 24,
+ "metadata": {
+ "id": "n9Qq7wcOU4X9"
+ },
+ "outputs": [],
"source": [
"async def run_async_function_call(self, generator, tool_manager):\n",
" answers = []\n",
@@ -725,12 +727,21 @@
" f\"Failed to execute the function for query: {query}, func: {result.data}, error: {e}\"\n",
" )\n",
" return None"
- ],
- "metadata": {
- "id": "n9Qq7wcOU4X9"
- },
- "execution_count": 24,
- "outputs": []
+ ]
+ }
+ ],
+ "metadata": {
+ "colab": {
+ "provenance": []
+ },
+ "kernelspec": {
+ "display_name": "Python 3",
+ "name": "python3"
+ },
+ "language_info": {
+ "name": "python"
}
- ]
+ },
+ "nbformat": 4,
+ "nbformat_minor": 0
}
diff --git a/notebooks/tutorials/adalflow_logger.ipynb b/notebooks/tutorials/adalflow_logger.ipynb
index ae5a7d83..64fd0114 100644
--- a/notebooks/tutorials/adalflow_logger.ipynb
+++ b/notebooks/tutorials/adalflow_logger.ipynb
@@ -1,21 +1,10 @@
{
- "nbformat": 4,
- "nbformat_minor": 0,
- "metadata": {
- "colab": {
- "provenance": []
- },
- "kernelspec": {
- "name": "python3",
- "display_name": "Python 3"
- },
- "language_info": {
- "name": "python"
- }
- },
"cells": [
{
"cell_type": "markdown",
+ "metadata": {
+ "id": "lLGpv1fLLIjF"
+ },
"source": [
"# Adalflow RAG Playbook example\n",
"\n",
@@ -26,10 +15,7 @@
"- RAG with dynamic data access and caching the embedding dynamically in a local storage.\n",
"\n",
"Here we will have have a look at an example with a local DB using FAISS"
- ],
- "metadata": {
- "id": "lLGpv1fLLIjF"
- }
+ ]
},
{
"cell_type": "code",
@@ -48,20 +34,18 @@
},
{
"cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
"source": [
- "import os\n",
- "from getpass import getpass\n",
- "\n",
- "# Prompt user to enter their API keys securely\n",
- "openai_api_key = getpass(\"Please enter your OpenAI API key: \")\n",
- "groq_api_key = getpass(\"Please enter your GROQ API key: \")\n",
- "\n",
- "# Set environment variables\n",
- "os.environ[\"OPENAI_API_KEY\"] = openai_api_key\n",
- "os.environ[\"GROQ_API_KEY\"] = groq_api_key\n",
- "\n",
- "print(\"API keys have been set.\")"
- ],
+ "!pip uninstall httpx anyio -y\n",
+ "!pip install \"anyio>=3.1.0,<4.0\"\n",
+ "!pip install httpx==0.24.1"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
@@ -69,21 +53,37 @@
"id": "-4c_AGBt3PlR",
"outputId": "275b050a-ce64-4b40-a5f9-4ccc12d92add"
},
- "execution_count": 2,
"outputs": [
{
- "output_type": "stream",
"name": "stdout",
+ "output_type": "stream",
"text": [
"Please enter your OpenAI API key: ยทยทยทยทยทยทยทยทยทยท\n",
"Please enter your GROQ API key: ยทยทยทยทยทยทยทยทยทยท\n",
"API keys have been set.\n"
]
}
+ ],
+ "source": [
+ "import os\n",
+ "from getpass import getpass\n",
+ "\n",
+ "# Prompt user to enter their API keys securely\n",
+ "openai_api_key = getpass(\"Please enter your OpenAI API key: \")\n",
+ "groq_api_key = getpass(\"Please enter your GROQ API key: \")\n",
+ "\n",
+ "# Set environment variables\n",
+ "os.environ[\"OPENAI_API_KEY\"] = openai_api_key\n",
+ "os.environ[\"GROQ_API_KEY\"] = groq_api_key\n",
+ "\n",
+ "print(\"API keys have been set.\")"
]
},
{
"cell_type": "markdown",
+ "metadata": {
+ "id": "4NztjiLR_EQE"
+ },
"source": [
"## Design\n",
"\n",
@@ -96,45 +96,38 @@
"2. Additionally, as we canโt always control the outputs of generators, we will provide customized logger and tracers(drop-in decorators) for them, for which we will explain in Tracing. This will not break the first objective.\n",
"\n",
"In the future, when we have more complex requirements from users, we will consider adding hooks/callbacks but we will do it in a way to keep the functional and user-facing APIs clean."
- ],
- "metadata": {
- "id": "4NztjiLR_EQE"
- }
+ ]
},
{
"cell_type": "code",
+ "execution_count": 3,
+ "metadata": {
+ "id": "d2H1vYoC_F-g"
+ },
+ "outputs": [],
"source": [
"import logging\n",
"\n",
"log = logging.getLogger(__name__)"
- ],
- "metadata": {
- "id": "d2H1vYoC_F-g"
- },
- "execution_count": 3,
- "outputs": []
+ ]
},
{
"cell_type": "code",
+ "execution_count": 4,
+ "metadata": {
+ "id": "e2GxAapG_TJH"
+ },
+ "outputs": [],
"source": [
"from adalflow.utils.logger import get_logger\n",
"\n",
"\n",
"root_logger = get_logger()"
- ],
- "metadata": {
- "id": "e2GxAapG_TJH"
- },
- "execution_count": 4,
- "outputs": []
+ ]
},
{
"cell_type": "code",
- "source": [
- "from adalflow.utils.logger import printc\n",
- "\n",
- "printc(\"All logging examples are done. Feeling green!\", color=\"green\")"
- ],
+ "execution_count": 5,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
@@ -142,30 +135,39 @@
"id": "Yk4oiBFE_asG",
"outputId": "470e30dc-1b31-40c1-9e48-30754ae54b45"
},
- "execution_count": 5,
"outputs": [
{
- "output_type": "stream",
"name": "stdout",
+ "output_type": "stream",
"text": [
"\u001b[32m2024-11-28 13:39:41 - [:3:] - All logging examples are done. Feeling green!\u001b[0m\n"
]
}
+ ],
+ "source": [
+ "from adalflow.utils.logger import printc\n",
+ "\n",
+ "printc(\"All logging examples are done. Feeling green!\", color=\"green\")"
]
},
{
"cell_type": "markdown",
+ "metadata": {
+ "id": "B8lmlT_9_nVP"
+ },
"source": [
"Set up all logs in one file\n",
"\n",
"Assume your source code is at src/task.py. You can log simply by:"
- ],
- "metadata": {
- "id": "B8lmlT_9_nVP"
- }
+ ]
},
{
"cell_type": "code",
+ "execution_count": 6,
+ "metadata": {
+ "id": "o_Ru1myM_c-J"
+ },
+ "outputs": [],
"source": [
"import logging\n",
"\n",
@@ -175,26 +177,11 @@
"class Task:\n",
" def __init__(self):\n",
" log.info(\"This is a user program child logger\")"
- ],
- "metadata": {
- "id": "o_Ru1myM_c-J"
- },
- "execution_count": 6,
- "outputs": []
+ ]
},
{
"cell_type": "code",
- "source": [
- "import logging\n",
- "from adalflow.utils.logger import get_logger\n",
- "\n",
- "root_logger = get_logger(level=\"DEBUG\", save_dir=\"./logs\") # log to ./logs/lib.log\n",
- "\n",
- "# run code from the library components such as generator\n",
- "# ....\n",
- "\n",
- "root_logger.info(\"This is the log in the main file\")"
- ],
+ "execution_count": 7,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
@@ -202,28 +189,43 @@
"id": "o7YPjEZk_ehg",
"outputId": "ad0f58e9-6f5c-4d00-e737-2fa1ad5ebd85"
},
- "execution_count": 7,
"outputs": [
{
- "output_type": "stream",
"name": "stdout",
+ "output_type": "stream",
"text": [
"2024-11-28 13:39:46 - - INFO - [:9:| ] - This is the log in the main file\n"
]
}
+ ],
+ "source": [
+ "import logging\n",
+ "from adalflow.utils.logger import get_logger\n",
+ "\n",
+ "root_logger = get_logger(level=\"DEBUG\", save_dir=\"./logs\") # log to ./logs/lib.log\n",
+ "\n",
+ "# run code from the library components such as generator\n",
+ "# ....\n",
+ "\n",
+ "root_logger.info(\"This is the log in the main file\")"
]
},
{
"cell_type": "markdown",
- "source": [
- "Separate library and application logs"
- ],
"metadata": {
"id": "Db1_Ob3X_gpe"
- }
+ },
+ "source": [
+ "Separate library and application logs"
+ ]
},
{
"cell_type": "code",
+ "execution_count": 8,
+ "metadata": {
+ "id": "rQWuFnUc_gNm"
+ },
+ "outputs": [],
"source": [
"from adalflow.utils.logger import get_logger\n",
"\n",
@@ -235,12 +237,21 @@
"class Task:\n",
" def __init__(self):\n",
" app_logger.info(\"This is a user program child logger\")"
- ],
- "metadata": {
- "id": "rQWuFnUc_gNm"
- },
- "execution_count": 8,
- "outputs": []
+ ]
}
- ]
+ ],
+ "metadata": {
+ "colab": {
+ "provenance": []
+ },
+ "kernelspec": {
+ "display_name": "Python 3",
+ "name": "python3"
+ },
+ "language_info": {
+ "name": "python"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 0
}
diff --git a/notebooks/tutorials/adalflow_modelclient.ipynb b/notebooks/tutorials/adalflow_modelclient.ipynb
index 1674c69a..1a2b3aba 100644
--- a/notebooks/tutorials/adalflow_modelclient.ipynb
+++ b/notebooks/tutorials/adalflow_modelclient.ipynb
@@ -87,6 +87,17 @@
"clear_output()"
]
},
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "!pip uninstall httpx anyio -y\n",
+ "!pip install \"anyio>=3.1.0,<4.0\"\n",
+ "!pip install httpx==0.24.1"
+ ]
+ },
{
"cell_type": "markdown",
"metadata": {
diff --git a/notebooks/tutorials/adalflow_rag_optimization.ipynb b/notebooks/tutorials/adalflow_rag_optimization.ipynb
index 34d208bf..2dbcc579 100644
--- a/notebooks/tutorials/adalflow_rag_optimization.ipynb
+++ b/notebooks/tutorials/adalflow_rag_optimization.ipynb
@@ -1,21 +1,10 @@
{
- "nbformat": 4,
- "nbformat_minor": 0,
- "metadata": {
- "colab": {
- "provenance": []
- },
- "kernelspec": {
- "name": "python3",
- "display_name": "Python 3"
- },
- "language_info": {
- "name": "python"
- }
- },
"cells": [
{
"cell_type": "markdown",
+ "metadata": {
+ "id": "xHF95Kr4CzGq"
+ },
"source": [
"# ๐ค Welcome to AdalFlow!\n",
"## The PyTorch library to auto-optimize any LLM task pipelines\n",
@@ -44,13 +33,13 @@
"- Build the standard RAG with Retriever and Generator components.\n",
"\n",
"- Learn how to connect the output-input between components to enable auto-text-grad optimization."
- ],
- "metadata": {
- "id": "xHF95Kr4CzGq"
- }
+ ]
},
{
"cell_type": "markdown",
+ "metadata": {
+ "id": "Kof5M6DRaKhh"
+ },
"source": [
"\n",
"# Installation\n",
@@ -63,10 +52,7 @@
"2. Setup `openai` and `groq` API key in the environment variables\n",
"\n",
"You can choose to use different client. You can import the model client you prefer. We support `Anthropic`, `Cohere`, `Google`, `GROQ`, `OpenAI`, `Transformer` and more in development. We will use OpenAI here as an example.Please refer to our [full installation guide](https://adalflow.sylph.ai/get_started/installation.html)"
- ],
- "metadata": {
- "id": "Kof5M6DRaKhh"
- }
+ ]
},
{
"cell_type": "code",
@@ -84,8 +70,22 @@
"clear_output()"
]
},
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "!pip uninstall httpx anyio -y\n",
+ "!pip install \"anyio>=3.1.0,<4.0\"\n",
+ "!pip install httpx==0.24.1"
+ ]
+ },
{
"cell_type": "markdown",
+ "metadata": {
+ "id": "KapUyHMM07pJ"
+ },
"source": [
"## Set Environment Variables\n",
"\n",
@@ -94,27 +94,11 @@
"Note: for normal `.py` projects, follow our [official installation guide](https://lightrag.sylph.ai/get_started/installation.html).\n",
"\n",
"*Go to [OpenAI](https://platform.openai.com/docs/introduction) to get API keys if you don't already have.*"
- ],
- "metadata": {
- "id": "KapUyHMM07pJ"
- }
+ ]
},
{
"cell_type": "code",
- "source": [
- "import os\n",
- "\n",
- "from getpass import getpass\n",
- "\n",
- "# Prompt user to enter their API keys securely\n",
- "openai_api_key = getpass(\"Please enter your OpenAI API key: \")\n",
- "\n",
- "\n",
- "# Set environment variables\n",
- "os.environ[\"OPENAI_API_KEY\"] = openai_api_key\n",
- "\n",
- "print(\"API keys have been set.\")"
- ],
+ "execution_count": 3,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
@@ -122,20 +106,38 @@
"id": "ONfzF9Puzdd_",
"outputId": "5fc0cd30-9ae7-443a-c06c-31e9edeafd69"
},
- "execution_count": 3,
"outputs": [
{
- "output_type": "stream",
"name": "stdout",
+ "output_type": "stream",
"text": [
"Please enter your OpenAI API key: ยทยทยทยทยทยทยทยทยทยท\n",
"API keys have been set.\n"
]
}
+ ],
+ "source": [
+ "import os\n",
+ "\n",
+ "from getpass import getpass\n",
+ "\n",
+ "# Prompt user to enter their API keys securely\n",
+ "openai_api_key = getpass(\"Please enter your OpenAI API key: \")\n",
+ "\n",
+ "\n",
+ "# Set environment variables\n",
+ "os.environ[\"OPENAI_API_KEY\"] = openai_api_key\n",
+ "\n",
+ "print(\"API keys have been set.\")"
]
},
{
"cell_type": "code",
+ "execution_count": 20,
+ "metadata": {
+ "id": "aE3I05BqOmd7"
+ },
+ "outputs": [],
"source": [
"import dspy\n",
"import re\n",
@@ -150,15 +152,15 @@
"from adalflow.core.retriever import Retriever\n",
"from adalflow.core.component import fun_to_component\n",
"from adalflow.components.model_client.openai_client import OpenAIClient"
- ],
- "metadata": {
- "id": "aE3I05BqOmd7"
- },
- "execution_count": 20,
- "outputs": []
+ ]
},
{
"cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "id": "cqUUoua9fUxQ"
+ },
+ "outputs": [],
"source": [
"gpt_4o_model = {\n",
" \"model_client\": OpenAIClient(),\n",
@@ -175,15 +177,37 @@
" \"max_tokens\": 2000,\n",
" },\n",
"}"
- ],
- "metadata": {
- "id": "cqUUoua9fUxQ"
- },
- "execution_count": null,
- "outputs": []
+ ]
},
{
"cell_type": "code",
+ "execution_count": 22,
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "id": "0irHeHUkOmL8",
+ "outputId": "61f778a2-9ec1-4fda-daa2-bcc7f31baa78"
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "HotPotQAData(id='5a8b57f25542995d1e6f1371', question='Were Scott Derrickson and Ed Wood of the same nationality?', answer='yes', gold_titles=\"{'Scott Derrickson', 'Ed Wood'}\") \n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "HotPotQAData(id='5a8b57f25542995d1e6f1371', question='Were Scott Derrickson and Ed Wood of the same nationality?', answer='yes', gold_titles=\"{'Scott Derrickson', 'Ed Wood'}\")"
+ ]
+ },
+ "execution_count": 22,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
"source": [
"def load_datasets():\n",
"\n",
@@ -215,37 +239,15 @@
" answer=\"yes\",\n",
" gold_titles=\"{'Scott Derrickson', 'Ed Wood'}\",\n",
")"
- ],
- "metadata": {
- "colab": {
- "base_uri": "https://localhost:8080/"
- },
- "id": "0irHeHUkOmL8",
- "outputId": "61f778a2-9ec1-4fda-daa2-bcc7f31baa78"
- },
- "execution_count": 22,
- "outputs": [
- {
- "output_type": "stream",
- "name": "stdout",
- "text": [
- "HotPotQAData(id='5a8b57f25542995d1e6f1371', question='Were Scott Derrickson and Ed Wood of the same nationality?', answer='yes', gold_titles=\"{'Scott Derrickson', 'Ed Wood'}\") \n"
- ]
- },
- {
- "output_type": "execute_result",
- "data": {
- "text/plain": [
- "HotPotQAData(id='5a8b57f25542995d1e6f1371', question='Were Scott Derrickson and Ed Wood of the same nationality?', answer='yes', gold_titles=\"{'Scott Derrickson', 'Ed Wood'}\")"
- ]
- },
- "metadata": {},
- "execution_count": 22
- }
]
},
{
"cell_type": "code",
+ "execution_count": 23,
+ "metadata": {
+ "id": "ZZIEtZYHNVjo"
+ },
+ "outputs": [],
"source": [
"class DspyRetriever(adal.Retriever):\n",
" def __init__(self, top_k: int = 3):\n",
@@ -474,25 +476,34 @@
" trainer.diagnose(dataset=trainset, split=\"train\")\n",
" # trainer.diagnose(dataset=valset, split=\"val\")\n",
" # trainer.diagnose(dataset=testset, split=\"test\")"
- ],
- "metadata": {
- "id": "ZZIEtZYHNVjo"
- },
- "execution_count": 23,
- "outputs": []
+ ]
},
{
"cell_type": "markdown",
+ "metadata": {
+ "id": "AmkbyxmuruUu"
+ },
"source": [
"# Issues and feedback\n",
"\n",
"If you encounter any issues, please report them here: [GitHub Issues](https://github.com/SylphAI-Inc/LightRAG/issues).\n",
"\n",
"For feedback, you can use either the [GitHub discussions](https://github.com/SylphAI-Inc/LightRAG/discussions) or [Discord](https://discord.gg/ezzszrRZvT)."
- ],
- "metadata": {
- "id": "AmkbyxmuruUu"
- }
+ ]
}
- ]
+ ],
+ "metadata": {
+ "colab": {
+ "provenance": []
+ },
+ "kernelspec": {
+ "display_name": "Python 3",
+ "name": "python3"
+ },
+ "language_info": {
+ "name": "python"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 0
}
diff --git a/notebooks/tutorials/adalflow_rag_playbook.ipynb b/notebooks/tutorials/adalflow_rag_playbook.ipynb
index 308ade6e..d4455474 100644
--- a/notebooks/tutorials/adalflow_rag_playbook.ipynb
+++ b/notebooks/tutorials/adalflow_rag_playbook.ipynb
@@ -1,21 +1,10 @@
{
- "nbformat": 4,
- "nbformat_minor": 0,
- "metadata": {
- "colab": {
- "provenance": []
- },
- "kernelspec": {
- "name": "python3",
- "display_name": "Python 3"
- },
- "language_info": {
- "name": "python"
- }
- },
"cells": [
{
"cell_type": "markdown",
+ "metadata": {
+ "id": "lLGpv1fLLIjF"
+ },
"source": [
"# Adalflow RAG Playbook example\n",
"\n",
@@ -26,10 +15,7 @@
"- RAG with dynamic data access and caching the embedding dynamically in a local storage.\n",
"\n",
"Here we will have have a look at an example with a local DB using FAISS"
- ],
- "metadata": {
- "id": "lLGpv1fLLIjF"
- }
+ ]
},
{
"cell_type": "code",
@@ -48,20 +34,18 @@
},
{
"cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
"source": [
- "import os\n",
- "from getpass import getpass\n",
- "\n",
- "# Prompt user to enter their API keys securely\n",
- "openai_api_key = getpass(\"Please enter your OpenAI API key: \")\n",
- "groq_api_key = getpass(\"Please enter your GROQ API key: \")\n",
- "\n",
- "# Set environment variables\n",
- "os.environ[\"OPENAI_API_KEY\"] = openai_api_key\n",
- "os.environ[\"GROQ_API_KEY\"] = groq_api_key\n",
- "\n",
- "print(\"API keys have been set.\")"
- ],
+ "!pip uninstall httpx anyio -y\n",
+ "!pip install \"anyio>=3.1.0,<4.0\"\n",
+ "!pip install httpx==0.24.1"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
@@ -69,21 +53,39 @@
"id": "-4c_AGBt3PlR",
"outputId": "a36f157b-0b18-4f3d-d5a8-09aa94743922"
},
- "execution_count": 2,
"outputs": [
{
- "output_type": "stream",
"name": "stdout",
+ "output_type": "stream",
"text": [
"Please enter your OpenAI API key: ยทยทยทยทยทยทยทยทยทยท\n",
"Please enter your GROQ API key: ยทยทยทยทยทยทยทยทยทยท\n",
"API keys have been set.\n"
]
}
+ ],
+ "source": [
+ "import os\n",
+ "from getpass import getpass\n",
+ "\n",
+ "# Prompt user to enter their API keys securely\n",
+ "openai_api_key = getpass(\"Please enter your OpenAI API key: \")\n",
+ "groq_api_key = getpass(\"Please enter your GROQ API key: \")\n",
+ "\n",
+ "# Set environment variables\n",
+ "os.environ[\"OPENAI_API_KEY\"] = openai_api_key\n",
+ "os.environ[\"GROQ_API_KEY\"] = groq_api_key\n",
+ "\n",
+ "print(\"API keys have been set.\")"
]
},
{
"cell_type": "code",
+ "execution_count": 4,
+ "metadata": {
+ "id": "V9LsGDnm3RbV"
+ },
+ "outputs": [],
"source": [
"from typing import Any, List, Optional\n",
"import os\n",
@@ -99,15 +101,15 @@
" TextSplitter,\n",
")\n",
"from adalflow.utils.global_config import get_adalflow_default_root_path"
- ],
- "metadata": {
- "id": "V9LsGDnm3RbV"
- },
- "execution_count": 4,
- "outputs": []
+ ]
},
{
"cell_type": "code",
+ "execution_count": 5,
+ "metadata": {
+ "id": "kWGTZxrw3Tli"
+ },
+ "outputs": [],
"source": [
"configs = {\n",
" \"embedder\": {\n",
@@ -135,15 +137,15 @@
" \"chunk_overlap\": 200,\n",
" },\n",
"}"
- ],
- "metadata": {
- "id": "kWGTZxrw3Tli"
- },
- "execution_count": 5,
- "outputs": []
+ ]
},
{
"cell_type": "code",
+ "execution_count": 6,
+ "metadata": {
+ "id": "1QE0PCKs4BLz"
+ },
+ "outputs": [],
"source": [
"def prepare_data_pipeline():\n",
" splitter = TextSplitter(**configs[\"text_splitter\"])\n",
@@ -172,15 +174,15 @@
" data_transformer = prepare_data_pipeline()\n",
" db.transform(data_transformer, key=\"data_transformer\")\n",
" db.save_state(index_path)"
- ],
- "metadata": {
- "id": "1QE0PCKs4BLz"
- },
- "execution_count": 6,
- "outputs": []
+ ]
},
{
"cell_type": "code",
+ "execution_count": 7,
+ "metadata": {
+ "id": "6Mu1HXhy4DIG"
+ },
+ "outputs": [],
"source": [
"RAG_PROMPT_TEMPLATE = r\"\"\"\n",
"{{task_desc}}\n",
@@ -284,42 +286,11 @@
" print(f\"context_str: \\n {context_str}\")\n",
"\n",
" return self.generate(query, context=context_str)"
- ],
- "metadata": {
- "id": "6Mu1HXhy4DIG"
- },
- "execution_count": 7,
- "outputs": []
+ ]
},
{
"cell_type": "code",
- "source": [
- "# Prepare initial documents\n",
- "doc1 = Document(\n",
- " meta_data={\"title\": \"Li Yin's profile\"},\n",
- " text=\"My name is Li Yin, I love rock climbing\" + \"lots of nonsense text\" * 500,\n",
- " id=\"doc1\",\n",
- ")\n",
- "doc2 = Document(\n",
- " meta_data={\"title\": \"Interviewing Li Yin\"},\n",
- " text=\"lots of more nonsense text\" * 250\n",
- " + \"Li Yin is an AI researcher and a software engineer\"\n",
- " + \"lots of more nonsense text\" * 250,\n",
- " id=\"doc2\",\n",
- ")\n",
- "\n",
- "# Prepare the database (only runs once)\n",
- "prepare_database_with_index([doc1, doc2], index_file=\"index.faiss\")\n",
- "\n",
- "# Initialize RAG\n",
- "rag = RAG(index_file=\"index.faiss\")\n",
- "print(rag)\n",
- "\n",
- "# Query the RAG system\n",
- "query = \"What is Li Yin's hobby and profession?\"\n",
- "response = rag.call(query)\n",
- "print(f\"Response: {response}\")"
- ],
+ "execution_count": 8,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
@@ -327,11 +298,10 @@
"id": "sPnx4PY34D1j",
"outputId": "f66d6f1a-70bf-40e9-a160-591fcfdcbed3"
},
- "execution_count": 8,
"outputs": [
{
- "output_type": "stream",
"name": "stderr",
+ "output_type": "stream",
"text": [
"Splitting Documents in Batches: 100%|โโโโโโโโโโ| 1/1 [00:00<00:00, 109.58it/s]\n",
"Batch embedding documents: 100%|โโโโโโโโโโ| 1/1 [00:01<00:00, 1.33s/it]\n",
@@ -339,8 +309,8 @@
]
},
{
- "output_type": "stream",
"name": "stdout",
+ "output_type": "stream",
"text": [
"Saved the state of the DB to /root/.adalflow/index.faiss\n",
"RAG(\n",
@@ -435,33 +405,38 @@
"Response: (GeneratorOutput(id=None, data={'answer': \"Li Yin's hobby is rock climbing and profession is an AI researcher and a software engineer.\"}, error=None, usage=CompletionUsage(completion_tokens=25, prompt_tokens=2713, total_tokens=2738), raw_response='{\\n \"answer\": \"Li Yin\\'s hobby is rock climbing and profession is an AI researcher and a software engineer.\"\\n}', metadata=None), ' My name is Li Yin, I love rock climbinglots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textLi Yin is an AI researcher and a software engineerlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more textLi Yin is an AI researcher and a software engineerlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense ')\n"
]
}
- ]
- },
- {
- "cell_type": "code",
+ ],
"source": [
- "# Add more documents at runtime\n",
- "doc3 = Document(\n",
- " meta_data={\"title\": \"Apple's profile\"},\n",
- " text=\"Apple is a cute dog with black and tan fur\" + \"lots of nonsense text\" * 500,\n",
- " id=\"doc3\",\n",
+ "# Prepare initial documents\n",
+ "doc1 = Document(\n",
+ " meta_data={\"title\": \"Li Yin's profile\"},\n",
+ " text=\"My name is Li Yin, I love rock climbing\" + \"lots of nonsense text\" * 500,\n",
+ " id=\"doc1\",\n",
")\n",
- "doc4 = Document(\n",
- " meta_data={\"title\": \"Apple's characteristics\"},\n",
+ "doc2 = Document(\n",
+ " meta_data={\"title\": \"Interviewing Li Yin\"},\n",
" text=\"lots of more nonsense text\" * 250\n",
- " + \"Apple is energetic, loves to play with her monkey toy\"\n",
+ " + \"Li Yin is an AI researcher and a software engineer\"\n",
" + \"lots of more nonsense text\" * 250,\n",
- " id=\"doc4\",\n",
+ " id=\"doc2\",\n",
")\n",
"\n",
- "rag.add_documents([doc3, doc4])\n",
- "rag.prepare_retriever()\n",
+ "# Prepare the database (only runs once)\n",
+ "prepare_database_with_index([doc1, doc2], index_file=\"index.faiss\")\n",
"\n",
- "# Test a new query\n",
- "query = \"What is Apple's favorite toy?\"\n",
+ "# Initialize RAG\n",
+ "rag = RAG(index_file=\"index.faiss\")\n",
+ "print(rag)\n",
+ "\n",
+ "# Query the RAG system\n",
+ "query = \"What is Li Yin's hobby and profession?\"\n",
"response = rag.call(query)\n",
"print(f\"Response: {response}\")"
- ],
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
@@ -469,11 +444,10 @@
"id": "bcC1-dCheVEC",
"outputId": "133bab3f-ff2e-40db-99dc-71d64af6283f"
},
- "execution_count": 9,
"outputs": [
{
- "output_type": "stream",
"name": "stderr",
+ "output_type": "stream",
"text": [
"Splitting Documents in Batches: 100%|โโโโโโโโโโ| 1/1 [00:00<00:00, 114.76it/s]\n",
"Batch embedding documents: 100%|โโโโโโโโโโ| 1/1 [00:00<00:00, 1.35it/s]\n",
@@ -481,25 +455,41 @@
]
},
{
- "output_type": "stream",
"name": "stdout",
+ "output_type": "stream",
"text": [
"Saved the state of the DB to /root/.adalflow/index.faiss\n",
"Response: (GeneratorOutput(id=None, data={'answer': \"Apple's favorite toy is her monkey toy.\"}, error=None, usage=CompletionUsage(completion_tokens=16, prompt_tokens=2647, total_tokens=2663), raw_response='{\\n \"answer\": \"Apple\\'s favorite toy is her monkey toy.\"\\n}', metadata=None), ' Apple is a cute dog with black and tan furlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots of nonsense textlots textApple is energetic, loves to play with her monkey toylots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textApple is energetic, loves to play with her monkey toylots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textLi Yin is an AI researcher and a software engineerlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more textLi Yin is an AI researcher and a software engineerlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more ')\n"
]
}
+ ],
+ "source": [
+ "# Add more documents at runtime\n",
+ "doc3 = Document(\n",
+ " meta_data={\"title\": \"Apple's profile\"},\n",
+ " text=\"Apple is a cute dog with black and tan fur\" + \"lots of nonsense text\" * 500,\n",
+ " id=\"doc3\",\n",
+ ")\n",
+ "doc4 = Document(\n",
+ " meta_data={\"title\": \"Apple's characteristics\"},\n",
+ " text=\"lots of more nonsense text\" * 250\n",
+ " + \"Apple is energetic, loves to play with her monkey toy\"\n",
+ " + \"lots of more nonsense text\" * 250,\n",
+ " id=\"doc4\",\n",
+ ")\n",
+ "\n",
+ "rag.add_documents([doc3, doc4])\n",
+ "rag.prepare_retriever()\n",
+ "\n",
+ "# Test a new query\n",
+ "query = \"What is Apple's favorite toy?\"\n",
+ "response = rag.call(query)\n",
+ "print(f\"Response: {response}\")"
]
},
{
"cell_type": "code",
- "source": [
- "# View all documents in the database\n",
- "print(\"All documents in the database:\")\n",
- "for item in rag.db.items:\n",
- " print(\n",
- " f\"ID: {item.id}, Title: {item.meta_data['title']}, Text: {item.text[:100]}...\"\n",
- " )"
- ],
+ "execution_count": 10,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
@@ -507,11 +497,10 @@
"id": "o9TzVv5GeZZ2",
"outputId": "bde56355-186c-4013-d702-b4530f82881b"
},
- "execution_count": 10,
"outputs": [
{
- "output_type": "stream",
"name": "stdout",
+ "output_type": "stream",
"text": [
"All documents in the database:\n",
"ID: doc1, Title: Li Yin's profile, Text: My name is Li Yin, I love rock climbinglots of nonsense textlots of nonsense textlots of nonsense te...\n",
@@ -520,7 +509,29 @@
"ID: doc4, Title: Apple's characteristics, Text: lots of more nonsense textlots of more nonsense textlots of more nonsense textlots of more nonsense ...\n"
]
}
+ ],
+ "source": [
+ "# View all documents in the database\n",
+ "print(\"All documents in the database:\")\n",
+ "for item in rag.db.items:\n",
+ " print(\n",
+ " f\"ID: {item.id}, Title: {item.meta_data['title']}, Text: {item.text[:100]}...\"\n",
+ " )"
]
}
- ]
+ ],
+ "metadata": {
+ "colab": {
+ "provenance": []
+ },
+ "kernelspec": {
+ "display_name": "Python 3",
+ "name": "python3"
+ },
+ "language_info": {
+ "name": "python"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 0
}
diff --git a/notebooks/tutorials/adalflow_text_splitter.ipynb b/notebooks/tutorials/adalflow_text_splitter.ipynb
index 4008f45a..7d5a05bd 100644
--- a/notebooks/tutorials/adalflow_text_splitter.ipynb
+++ b/notebooks/tutorials/adalflow_text_splitter.ipynb
@@ -11,6 +11,17 @@
"!pip install adalflow[openai,groq,faiss-cpu]"
]
},
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "!pip uninstall httpx anyio -y\n",
+ "!pip install \"anyio>=3.1.0,<4.0\"\n",
+ "!pip install httpx==0.24.1"
+ ]
+ },
{
"cell_type": "code",
"execution_count": null,
diff --git a/notebooks/tutorials/adalflow_tracing.ipynb b/notebooks/tutorials/adalflow_tracing.ipynb
index ef3d2b25..fbb6f9ce 100644
--- a/notebooks/tutorials/adalflow_tracing.ipynb
+++ b/notebooks/tutorials/adalflow_tracing.ipynb
@@ -1,31 +1,17 @@
{
- "nbformat": 4,
- "nbformat_minor": 0,
- "metadata": {
- "colab": {
- "provenance": []
- },
- "kernelspec": {
- "name": "python3",
- "display_name": "Python 3"
- },
- "language_info": {
- "name": "python"
- }
- },
"cells": [
{
"cell_type": "markdown",
+ "metadata": {
+ "id": "lLGpv1fLLIjF"
+ },
"source": [
"# Tracing\n",
"\n",
"In particular, we provide two tracing methods to help you develop and improve the Generator:\n",
"\n",
"1. Trace the history change(states) on prompt during your development process. Developers typically go through a long process of prompt optimization and it is frustrating to lose track of the prompt changes when your current change actually makes the performance much worse.\n"
- ],
- "metadata": {
- "id": "lLGpv1fLLIjF"
- }
+ ]
},
{
"cell_type": "code",
@@ -44,20 +30,18 @@
},
{
"cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
"source": [
- "import os\n",
- "from getpass import getpass\n",
- "\n",
- "# Prompt user to enter their API keys securely\n",
- "openai_api_key = getpass(\"Please enter your OpenAI API key: \")\n",
- "groq_api_key = getpass(\"Please enter your GROQ API key: \")\n",
- "\n",
- "# Set environment variables\n",
- "os.environ[\"OPENAI_API_KEY\"] = openai_api_key\n",
- "os.environ[\"GROQ_API_KEY\"] = groq_api_key\n",
- "\n",
- "print(\"API keys have been set.\")"
- ],
+ "!pip uninstall httpx anyio -y\n",
+ "!pip install \"anyio>=3.1.0,<4.0\"\n",
+ "!pip install httpx==0.24.1"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
@@ -65,30 +49,48 @@
"id": "-4c_AGBt3PlR",
"outputId": "85aba038-ee9c-463d-bdbd-027cbfff0094"
},
- "execution_count": 2,
"outputs": [
{
- "output_type": "stream",
"name": "stdout",
+ "output_type": "stream",
"text": [
"Please enter your OpenAI API key: ยทยทยทยทยทยทยทยทยทยท\n",
"Please enter your GROQ API key: ยทยทยทยทยทยทยทยทยทยท\n",
"API keys have been set.\n"
]
}
+ ],
+ "source": [
+ "import os\n",
+ "from getpass import getpass\n",
+ "\n",
+ "# Prompt user to enter their API keys securely\n",
+ "openai_api_key = getpass(\"Please enter your OpenAI API key: \")\n",
+ "groq_api_key = getpass(\"Please enter your GROQ API key: \")\n",
+ "\n",
+ "# Set environment variables\n",
+ "os.environ[\"OPENAI_API_KEY\"] = openai_api_key\n",
+ "os.environ[\"GROQ_API_KEY\"] = groq_api_key\n",
+ "\n",
+ "print(\"API keys have been set.\")"
]
},
{
"cell_type": "markdown",
- "source": [
- "We created a GeneratorStateLogger to handle the logging and saving into json files. To further simplify developersโs process, we provides a class decorator trace_generator_states where a single line of code can be added to any of your task component. It will automatically track any attributes of type Generator."
- ],
"metadata": {
"id": "yWi2uEiE6UIf"
- }
+ },
+ "source": [
+ "We created a GeneratorStateLogger to handle the logging and saving into json files. To further simplify developersโs process, we provides a class decorator trace_generator_states where a single line of code can be added to any of your task component. It will automatically track any attributes of type Generator."
+ ]
},
{
"cell_type": "code",
+ "execution_count": 13,
+ "metadata": {
+ "id": "qk9pkcCVzdek"
+ },
+ "outputs": [],
"source": [
"from adalflow.tracing import trace_generator_states\n",
"from adalflow.core import Component, Generator\n",
@@ -110,33 +112,33 @@
"\n",
" def call(self, query: str) -> str:\n",
" return self.doc(prompt_kwargs={\"input_str\": query}).data"
- ],
- "metadata": {
- "id": "qk9pkcCVzdek"
- },
- "execution_count": 13,
- "outputs": []
+ ]
},
{
"cell_type": "markdown",
- "source": [
- "Here is the folder structer of where the trace is generated as a .json file and also an example output below"
- ],
"metadata": {
"id": "LAZUSnYn-lnI"
- }
+ },
+ "source": [
+ "Here is the folder structer of where the trace is generated as a .json file and also an example output below"
+ ]
},
{
"cell_type": "markdown",
- "source": [
- ""
- ],
"metadata": {
"id": "cVofNXVW-EMo"
- }
+ },
+ "source": [
+ ""
+ ]
},
{
"cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "id": "dPd9i6_t7ERJ"
+ },
+ "outputs": [],
"source": [
"\"\"\"\n",
" {\n",
@@ -173,12 +175,21 @@
" ]\n",
"}\n",
"\"\"\""
- ],
- "metadata": {
- "id": "dPd9i6_t7ERJ"
- },
- "execution_count": null,
- "outputs": []
+ ]
}
- ]
+ ],
+ "metadata": {
+ "colab": {
+ "provenance": []
+ },
+ "kernelspec": {
+ "display_name": "Python 3",
+ "name": "python3"
+ },
+ "language_info": {
+ "name": "python"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 0
}
| | |