Add input validation notebook

Author: irgolic

docs/examples/input_validation.ipynb

@@ -0,0 +1,123 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Input Validation\n",
+    "\n",
+    "Guardrails supports validating inputs (prompts, instructions, msg_history) with string validators."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "In XML, specify the validators on the `prompt` or `instructions` tag, as such:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "is_executing": true
+   },
+   "outputs": [],
+   "source": [
+    "rail_spec = \"\"\"\n",
+    "<rail version=\"0.1\">\n",
+    "<prompt\n",
+    "    validators=\"two-words\"\n",
+    "    on-fail-two-words=\"exception\"\n",
+    ">\n",
+    "This is not two words\n",
+    "</prompt>\n",
+    "<output type=\"string\">\n",
+    "</output>\n",
+    "</rail>\n",
+    "\"\"\"\n",
+    "\n",
+    "from guardrails import Guard\n",
+    "guard = Guard.from_rail_string(rail_spec)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "When `fix` is specified as the on-fail handler, the prompt will automatically be amended before calling the LLM.\n",
+    "\n",
+    "In any other case (for example, `exception`), a `ValidationException` will be returned in the outcome."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "is_executing": true
+   },
+   "outputs": [],
+   "source": [
+    "import openai\n",
+    "\n",
+    "outcome = guard(\n",
+    "    openai.ChatCompletion.create,\n",
+    ")\n",
+    "outcome.error"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "When using pydantic to initialize a `Guard`, input validators can be specified by composition, as such:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from guardrails.validators import TwoWords\n",
+    "from pydantic import BaseModel\n",
+    "\n",
+    "\n",
+    "class Pet(BaseModel):\n",
+    "    name: str\n",
+    "    age: int\n",
+    "\n",
+    "\n",
+    "guard = Guard.from_pydantic(Pet)\n",
+    "guard.with_prompt_validation([TwoWords(on_fail=\"exception\")])\n",
+    "\n",
+    "outcome = guard(\n",
+    "    openai.ChatCompletion.create,\n",
+    "    prompt=\"This is not two words\",\n",
+    ")\n",
+    "outcome.error"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.11.0"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}