wehub-resource-sync
a0c8464e58
Build Package / build (ubuntu-latest) (push) Failing after 1s
CodeQL / Analyze (python) (push) Failing after 1s
Core Typecheck / core-typecheck (push) Failing after 1s
Linting / lint (push) Failing after 1s
llama-dev tests / test-llama-dev (push) Failing after 1s
Publish Sub-Package to PyPI if Needed / publish_subpackage_if_needed (push) Has been skipped
Sync Docs to Developer Hub / sync-docs (push) Failing after 0s
Build Package / build (windows-latest) (push) Has been cancelled
227 lines
15 KiB
Plaintext
227 lines
15 KiB
Plaintext
{
|
|
"cells": [
|
|
{
|
|
"attachments": {},
|
|
"cell_type": "markdown",
|
|
"id": "abbcf063",
|
|
"metadata": {},
|
|
"source": [
|
|
"<a href=\"https://colab.research.google.com/github/run-llama/llama_index/blob/main/docs/examples/output_parsing/guidance_pydantic_program.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"id": "530c973e-916d-4c9e-9365-e2d5306d7e3d",
|
|
"metadata": {},
|
|
"source": [
|
|
"# Guidance Pydantic Program"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"id": "18461ba1-6978-4b5b-861e-6dceec36857b",
|
|
"metadata": {},
|
|
"source": [
|
|
"Generate structured data with [**guidance**](https://github.com/microsoft/guidance) via LlamaIndex. \n",
|
|
"\n",
|
|
"\n",
|
|
"With guidance, you can guarantee the output structure is correct by *forcing* the LLM to output desired tokens. \n",
|
|
"This is especialy helpful when you are using lower-capacity model (e.g. the current open source models), which otherwise would struggle to generate valid output that fits the desired output schema."
|
|
]
|
|
},
|
|
{
|
|
"attachments": {},
|
|
"cell_type": "markdown",
|
|
"id": "c30252a1",
|
|
"metadata": {},
|
|
"source": [
|
|
"If you're opening this Notebook on colab, you will probably need to install LlamaIndex 🦙."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"id": "5c0a7972",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"%pip install llama-index-program-guidance"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"id": "faaf40a1",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"!pip install llama-index"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"id": "f7a83b49-5c34-45d5-8cf4-62f348fb1299",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"from pydantic import BaseModel\n",
|
|
"from typing import List\n",
|
|
"from guidance.llms import OpenAI\n",
|
|
"\n",
|
|
"from llama_index.program.guidance import GuidancePydanticProgram"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"id": "0563f1ba-8086-4dcc-ba35-bfda31c45ae4",
|
|
"metadata": {},
|
|
"source": [
|
|
"Define output schema"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"id": "42053ea8-2580-4639-9dcf-566e8427c44e",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"class Song(BaseModel):\n",
|
|
" title: str\n",
|
|
" length_seconds: int\n",
|
|
"\n",
|
|
"\n",
|
|
"class Album(BaseModel):\n",
|
|
" name: str\n",
|
|
" artist: str\n",
|
|
" songs: List[Song]"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"id": "4afff44e-a746-4b9f-85a9-72058bcdd29f",
|
|
"metadata": {},
|
|
"source": [
|
|
"Define guidance pydantic program"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"id": "fe756697-c299-4f9a-a108-944b6693f824",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"program = GuidancePydanticProgram(\n",
|
|
" output_cls=Album,\n",
|
|
" prompt_template_str=(\n",
|
|
" \"Generate an example album, with an artist and a list of songs. Using\"\n",
|
|
" \" the movie {{movie_name}} as inspiration\"\n",
|
|
" ),\n",
|
|
" guidance_llm=OpenAI(\"text-davinci-003\"),\n",
|
|
" verbose=True,\n",
|
|
")"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"id": "b7be01dc-433e-4485-bab0-36a04c3afbcb",
|
|
"metadata": {},
|
|
"source": [
|
|
"Run program to get structured output. \n",
|
|
"Text highlighted in blue is variables specified by us, text highlighted in green is generated by the LLM."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"id": "25d02228-2907-4810-932e-83ec9fc71f6b",
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"data": {
|
|
"text/html": [
|
|
"<div id=\"guidance-stop-button-c9591b41-7839-48fe-92c0-43860b22472d\" style=\"cursor: pointer; margin: 0px; display: none; float: right; padding: 3px; border-radius: 4px 4px 4px 4px; border: 0px solid rgba(127, 127, 127, 1); padding-left: 10px; padding-right: 10px; font-size: 13px; background-color: rgba(127, 127, 127, 0.25);\">Stop program</div><div id=\"guidance-content-c9591b41-7839-48fe-92c0-43860b22472d\"><pre style='margin: 0px; padding: 0px; padding-left: 8px; margin-left: -8px; border-radius: 0px; border-left: 1px solid rgba(127, 127, 127, 0.2); white-space: pre-wrap; font-family: ColfaxAI, Arial; font-size: 15px; line-height: 23px;'>Generate an example album, with an artist and a list of songs. Using the movie <span style='background-color: rgba(0, 138.56128016, 250.76166089, 0.25); display: inline;' title='{{movie_name}}'>The Shining</span> as inspiration\n",
|
|
"```json\n",
|
|
"{\n",
|
|
" "name": "<span style='background-color: rgba(0, 165, 0, 0.25); opacity: 1.0; display: inline;' title='{{gen 'name' stop='"'}}'>The Shining</span>",\n",
|
|
" "artist": "<span style='background-color: rgba(0, 165, 0, 0.25); opacity: 1.0; display: inline;' title='{{gen 'artist' stop='"'}}'>Jack Torrance</span>",\n",
|
|
" "songs": [<span style='opacity: 1.0; display: inline; background-color: rgba(165, 165, 165, 0.1);' title='{{#geneach 'songs' stop=']'}}{{#unless @first}}, {{/unless}}{\n",
|
|
" "title": "{{gen 'title' stop='"'}}",\n",
|
|
" "length_seconds": "{{gen 'length_seconds' stop='"'}}",\n",
|
|
"}{{/geneach}}'><span style='background-color: rgba(0, 138.56128016, 250.76166089, 0.25); display: inline;' title='{{#unless @first}}, {{/unless}}'></span>{\n",
|
|
" "title": "<span style='background-color: rgba(0, 165, 0, 0.25); opacity: 1.0; display: inline;' title='{{gen 'title' stop='"'}}'>All Work and No Play</span>",\n",
|
|
" "length_seconds": "<span style='background-color: rgba(0, 165, 0, 0.25); opacity: 1.0; display: inline;' title='{{gen 'length_seconds' stop='"'}}'>180</span>",\n",
|
|
"}<span style='background-color: rgba(0, 138.56128016, 250.76166089, 0.25); display: inline;' title='{{#unless @first}}, {{/unless}}'>, </span>{\n",
|
|
" "title": "<span style='background-color: rgba(0, 165, 0, 0.25); opacity: 1.0; display: inline;' title='{{gen 'title' stop='"'}}'>The Overlook Hotel</span>",\n",
|
|
" "length_seconds": "<span style='background-color: rgba(0, 165, 0, 0.25); opacity: 1.0; display: inline;' title='{{gen 'length_seconds' stop='"'}}'>240</span>",\n",
|
|
"}<span style='background-color: rgba(0, 138.56128016, 250.76166089, 0.25); display: inline;' title='{{#unless @first}}, {{/unless}}'>, </span>{\n",
|
|
" "title": "<span style='background-color: rgba(0, 165, 0, 0.25); opacity: 1.0; display: inline;' title='{{gen 'title' stop='"'}}'>The Shining</span>",\n",
|
|
" "length_seconds": "<span style='background-color: rgba(0, 165, 0, 0.25); opacity: 1.0; display: inline;' title='{{gen 'length_seconds' stop='"'}}'>210</span>",\n",
|
|
"}</span>],\n",
|
|
"}\n",
|
|
"```</pre></div>\n",
|
|
"<script type=\"text/javascript\">(()=>{var t={296:(t,e,n)=>{var i=NaN,o=\"[object Symbol]\",r=/^\\s+|\\s+$/g,a=/^[-+]0x[0-9a-f]+$/i,s=/^0b[01]+$/i,c=/^0o[0-7]+$/i,d=parseInt,u=\"object\"==typeof n.g&&n.g&&n.g.Object===Object&&n.g,l=\"object\"==typeof self&&self&&self.Object===Object&&self,f=u||l||Function(\"return this\")(),h=Object.prototype.toString,p=Math.max,m=Math.min,g=function(){return f.Date.now()};function b(t){var e=typeof t;return!!t&&(\"object\"==e||\"function\"==e)}function y(t){if(\"number\"==typeof t)return t;if(function(t){return\"symbol\"==typeof t||function(t){return!!t&&\"object\"==typeof t}(t)&&h.call(t)==o}(t))return i;if(b(t)){var e=\"function\"==typeof t.valueOf?t.valueOf():t;t=b(e)?e+\"\":e}if(\"string\"!=typeof t)return 0===t?t:+t;t=t.replace(r,\"\");var n=s.test(t);return n||c.test(t)?d(t.slice(2),n?2:8):a.test(t)?i:+t}t.exports=function(t,e,n){var i,o,r,a,s,c,d=0,u=!1,l=!1,f=!0;if(\"function\"!=typeof t)throw new TypeError(\"Expected a function\");function h(e){var n=i,r=o;return i=o=void 0,d=e,a=t.apply(r,n)}function v(t){var n=t-c;return void 0===c||n>=e||n<0||l&&t-d>=r}function _(){var t=g();if(v(t))return w(t);s=setTimeout(_,function(t){var n=e-(t-c);return l?m(n,r-(t-d)):n}(t))}function w(t){return s=void 0,f&&i?h(t):(i=o=void 0,a)}function j(){var t=g(),n=v(t);if(i=arguments,o=this,c=t,n){if(void 0===s)return function(t){return d=t,s=setTimeout(_,e),u?h(t):a}(c);if(l)return s=setTimeout(_,e),h(c)}return void 0===s&&(s=setTimeout(_,e)),a}return e=y(e)||0,b(n)&&(u=!!n.leading,r=(l=\"maxWait\"in n)?p(y(n.maxWait)||0,e):r,f=\"trailing\"in n?!!n.trailing:f),j.cancel=function(){void 0!==s&&clearTimeout(s),d=0,i=c=o=s=void 0},j.flush=function(){return void 0===s?a:w(g())},j}},777:t=>{var e,n,i=Math.max,o=(e=function(t,e){return function(t,e,n){if(\"function\"!=typeof t)throw new TypeError(\"Expected a function\");return setTimeout((function(){t.apply(void 0,n)}),1)}(t,0,e)},n=i(void 0===n?e.length-1:n,0),function(){for(var t=arguments,o=-1,r=i(t.length-n,0),a=Array(r);++o<r;)a[o]=t[n+o];o=-1;for(var s=Array(n+1);++o<n;)s[o]=t[o];return s[n]=a,function(t,e,n){switch(n.length){case 0:return t.call(e);case 1:return t.call(e,n[0]);case 2:return t.call(e,n[0],n[1]);case 3:return t.call(e,n[0],n[1],n[2])}return t.apply(e,n)}(e,this,s)});t.exports=o}},e={};function n(i){var o=e[i];if(void 0!==o)return o.exports;var r=e[i]={exports:{}};return t[i](r,r.exports,n),r.exports}n.n=t=>{var e=t&&t.__esModule?()=>t.default:()=>t;return n.d(e,{a:e}),e},n.d=(t,e)=>{for(var i in e)n.o(e,i)&&!n.o(t,i)&&Object.defineProperty(t,i,{enumerable:!0,get:e[i]})},n.g=function(){if(\"object\"==typeof globalThis)return globalThis;try{return this||new Function(\"return this\")()}catch(t){if(\"object\"==typeof window)return window}}(),n.o=(t,e)=>Object.prototype.hasOwnProperty.call(t,e),(()=>{\"use strict\";const t=t=>{const e=new Set;do{for(const n of Reflect.ownKeys(t))e.add([t,n])}while((t=Reflect.getPrototypeOf(t))&&t!==Object.prototype);return e};function e(e,{include:n,exclude:i}={}){const o=t=>{const e=e=>\"string\"==typeof e?t===e:e.test(t);return n?n.some(e):!i||!i.some(e)};for(const[n,i]of t(e.constructor.prototype)){if(\"constructor\"===i||!o(i))continue;const t=Reflect.getOwnPropertyDescriptor(n,i);t&&\"function\"==typeof t.value&&(e[i]=e[i].bind(e))}return e}var i=n(777),o=n.n(i),r=n(296),a=n.n(r);class s{constructor(t,n){e(this),this.interfaceId=t,this.callbackMap={},this.data={},this.pendingData={},this.jcomm=new c(\"guidance_interface_target_\"+this.interfaceId,this.updateData,\"open\"),this.debouncedSendPendingData500=a()(this.sendPendingData,500),this.debouncedSendPendingData1000=a()(this.sendPendingData,1e3),n&&o()(n)}send(t,e){this.addPendingData(t,e),this.sendPendingData()}sendEvent(t){for(const e of Object.keys(t))this.addPendingData(e,t[e]);this.sendPendingData()}debouncedSendEvent500(t){for(const e of Object.keys(t))this.addPendingData(e,t[e]);this.debouncedSendPendingData500()}debouncedSend500(t,e){this.addPendingData(t,e),this.debouncedSendPendingData500()}debouncedSend1000(t,e){this.addPendingData(t,e),this.debouncedSendPendingData1000()}addPendingData(t,e){Array.isArray(t)||(t=[t]);for(const n in t)this.pendingData[t[n]]=e}updateData(t){t=JSON.parse(t.data);for(const e in t)this.data[e]=t[e];for(const e in t)e in this.callbackMap&&this.callbackMap[e](this.data[e])}subscribe(t,e){this.callbackMap[t]=e,o()((e=>this.callbackMap[t](this.data[t])))}sendPendingData(){this.jcomm.send_data(this.pendingData),this.pendingData={}}}class c{constructor(t,e,n=\"open\"){this._fire_callback=this._fire_callback.bind(this),this._register=this._register.bind(this),this.jcomm=void 0,this.callback=e,void 0!==window.Jupyter?\"register\"===n?Jupyter.notebook.kernel.comm_manager.register_target(t,this._register):(this.jcomm=Jupyter.notebook.kernel.comm_manager.new_comm(t),this.jcomm.on_msg(this._fire_callback)):void 0!==window._mgr&&(\"register\"===n?window._mgr.widgetManager.proxyKernel.registerCommTarget(t,this._register):(this.jcomm=window._mgr.widgetManager.proxyKernel.createComm(t),this.jcomm.open({},\"\"),this.jcomm.onMsg=this._fire_callback))}send_data(t){void 0!==this.jcomm?this.jcomm.send(t):console.error(\"Jupyter comm module not yet loaded! So we can't send the message.\")}_register(t,e){this.jcomm=t,this.jcomm.on_msg(this._fire_callback)}_fire_callback(t){this.callback(t.content.data)}}class d{constructor(t,n){e(this),this.id=t,this.comm=new s(t),this.comm.subscribe(\"append\",this.appendData),this.comm.subscribe(\"replace\",this.replaceData),this.comm.subscribe(\"event\",this.eventOccurred),this.element=document.getElementById(\"guidance-content-\"+t),this.stop_button=document.getElementById(\"guidance-stop-button-\"+t),this.stop_button.onclick=()=>this.comm.send(\"event\",\"stop\")}appendData(t){t&&(this.stop_button.style.display=\"inline-block\",this.element.innerHTML+=t)}replaceData(t){t&&(this.stop_button.style.display=\"inline-block\",this.element.innerHTML=t)}eventOccurred(t){\"complete\"===t&&(this.stop_button.style.display=\"none\")}}window._guidanceDisplay=function(t,e){return new d(t,e)}})()})();; window._guidanceDisplay(\"c9591b41-7839-48fe-92c0-43860b22472d\");</script>"
|
|
]
|
|
},
|
|
"metadata": {},
|
|
"output_type": "display_data"
|
|
}
|
|
],
|
|
"source": [
|
|
"output = program(movie_name=\"The Shining\")"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"id": "27ec0777-28d5-494b-b419-daf6bce2b20e",
|
|
"metadata": {},
|
|
"source": [
|
|
"The output is a valid Pydantic object that we can then use to call functions/APIs. "
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"id": "3e51bcf4-e7df-47b9-b380-8e5b900a31e1",
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"data": {
|
|
"text/plain": [
|
|
"Album(name='The Shining', artist='Jack Torrance', songs=[Song(title='All Work and No Play', length_seconds=180), Song(title='The Overlook Hotel', length_seconds=240), Song(title='The Shining', length_seconds=210)])"
|
|
]
|
|
},
|
|
"execution_count": null,
|
|
"metadata": {},
|
|
"output_type": "execute_result"
|
|
}
|
|
],
|
|
"source": [
|
|
"output"
|
|
]
|
|
}
|
|
],
|
|
"metadata": {
|
|
"kernelspec": {
|
|
"display_name": "Python 3 (ipykernel)",
|
|
"language": "python",
|
|
"name": "python3"
|
|
},
|
|
"language_info": {
|
|
"codemirror_mode": {
|
|
"name": "ipython",
|
|
"version": 3
|
|
},
|
|
"file_extension": ".py",
|
|
"mimetype": "text/x-python",
|
|
"name": "python",
|
|
"nbconvert_exporter": "python",
|
|
"pygments_lexer": "ipython3"
|
|
}
|
|
},
|
|
"nbformat": 4,
|
|
"nbformat_minor": 5
|
|
}
|