refactor: openai function calling docs

langchain-ai · Oct 12, 2023 · 988ae44 · 988ae44
1 parent e83888e
commit 988ae44
Show file tree

Hide file tree

Showing 3 changed files with 123 additions and 69 deletions.
diff --git a/docs/extras/modules/model_io/models/chat/how_to/function_calling.mdx b/docs/extras/modules/model_io/models/chat/how_to/function_calling.mdx
@@ -17,10 +17,82 @@ directly to the model and call it, as shown below.
 
 ## Usage
 
+There are two main ways to apply functions to your OpenAI calls.
+
+The first and simplest is by attaching a function directly to the `.invoke({})` method:
+
+```typescript
+/* Define your function schema */
+const extractionFunctionSchema = {...}
+
+/* Instantiate ChatOpenAI class */
+const model = new ChatOpenAI({ modelName: "gpt-4" });
+
+/**
+ * Call the .invoke method on the model, directly passing
+ * the function arguments as call args.
+ */
+const result = await model.invoke([new HumanMessage("What a beautiful day!")], {
+  functions: [extractionFunctionSchema],
+  function_call: { name: "extractor" },
+});
+
+console.log({ result });
+```
+
+The second way is by directly binding the function to your model. Binding function arguments to your model is useful when you want to call the same function twice.
+Calling the `.bind({})` method attaches any call arguments passed in to all future calls to the model.
+
+```typescript
+/* Define your function schema */
+const extractionFunctionSchema = {...}
+
+/* Instantiate ChatOpenAI class and bind function arguments to the model */
+const model = new ChatOpenAI({ modelName: "gpt-4" }).bind({
+  functions: [extractionFunctionSchema],
+  function_call: { name: "extractor" },
+});
+
+/* Now we can call the model without having to pass the function arguments in again */
+const result = await model.invoke([new HumanMessage("What a beautiful day!")]);
+
+console.log({ result });
+```
+
 OpenAI requires parameter schemas in the format below, where `parameters` must be [JSON Schema](https://json-schema.org/).
-Specifying the `function_call` parameter will force the model to return a response using the specified function.
+When adding call arguments to your model, specifying the `function_call` argument will force the model to return a response using the specified function.
 This is useful if you have multiple schemas you'd like the model to pick from.
 
+Example function schema:
+
+```typescript
+const extractionFunctionSchema = {
+  name: "extractor",
+  description: "Extracts fields from the input.",
+  parameters: {
+    type: "object",
+    properties: {
+      tone: {
+        type: "string",
+        enum: ["positive", "negative"],
+        description: "The overall tone of the input",
+      },
+      word_count: {
+        type: "number",
+        description: "The number of words in the input",
+      },
+      chat_response: {
+        type: "string",
+        description: "A response to the human's input",
+      },
+    },
+    required: ["tone", "word_count", "chat_response"],
+  },
+};
+```
+
+Now to put it all together:
+
 <CodeBlock language="typescript">{OpenAIFunctionsExample}</CodeBlock>
 
 ## Usage with Zod

diff --git a/examples/src/models/chat/openai_functions.ts b/examples/src/models/chat/openai_functions.ts
@@ -25,10 +25,6 @@ const extractionFunctionSchema = {
   },
 };
 
-// Bind function arguments to the model.
-// All subsequent invoke calls will use the bound parameters.
-// "functions.parameters" must be formatted as JSON Schema
-// Omit "function_call" if you want the model to choose a function to call.
 const model = new ChatOpenAI({
   modelName: "gpt-4",
 }).bind({
@@ -39,34 +35,22 @@ const model = new ChatOpenAI({
 const result = await model.invoke([new HumanMessage("What a beautiful day!")]);
 
 console.log(result);
-
 /*
-  AIMessage {
-    content: '',
-    name: undefined,
-    additional_kwargs: {
-      function_call: {
-        name: 'extractor',
-        arguments: '{\n' +
-          '  "tone": "positive",\n' +
-          '  "word_count": 4,\n' +
-          '  "chat_response": "It certainly is a beautiful day!"\n' +
-          '}'
-      }
+AIMessage {
+  lc_serializable: true,
+  lc_kwargs: { content: '', additional_kwargs: { function_call: [Object] } },
+  lc_namespace: [ 'langchain', 'schema' ],
+  content: '',
+  name: undefined,
+  additional_kwargs: {
+    function_call: {
+      name: 'extractor',
+      arguments: '{\n' +
+        '  "tone": "positive",\n' +
+        '  "word_count": 4,\n' +
+        `  "chat_response": "I'm glad you're enjoying the day! What makes it so beautiful for you?"\n` +
+        '}'
     }
   }
-*/
-
-// Alternatively, you can pass function call arguments as an additional argument as a one-off:
-/*
-const model = new ChatOpenAI({
-  modelName: "gpt-4",
-});
-
-const result = await model.call([
-  new HumanMessage("What a beautiful day!")
-], {
-  functions: [extractionFunctionSchema],
-  function_call: {name: "extractor"}
-});
+}
 */
diff --git a/examples/src/models/chat/openai_functions_zod.ts b/examples/src/models/chat/openai_functions_zod.ts
@@ -3,54 +3,52 @@ import { HumanMessage } from "langchain/schema";
 import { z } from "zod";
 import { zodToJsonSchema } from "zod-to-json-schema";
 
-const extractionFunctionZodSchema = z.object({
-  tone: z
-    .enum(["positive", "negative"])
-    .describe("The overall tone of the input"),
-  entity: z.string().describe("The entity mentioned in the input"),
-  word_count: z.number().describe("The number of words in the input"),
-  chat_response: z.string().describe("A response to the human's input"),
-  final_punctuation: z
-    .optional(z.string())
-    .describe("The final punctuation mark in the input, if any."),
-});
+const extractionFunctionSchema = {
+  name: "extractor",
+  description: "Extracts fields from the input.",
+  parameters: zodToJsonSchema(
+    z.object({
+      tone: z
+        .enum(["positive", "negative"])
+        .describe("The overall tone of the input"),
+      entity: z.string().describe("The entity mentioned in the input"),
+      word_count: z.number().describe("The number of words in the input"),
+      chat_response: z.string().describe("A response to the human's input"),
+      final_punctuation: z
+        .optional(z.string())
+        .describe("The final punctuation mark in the input, if any."),
+    })
+  ),
+};
 
-// Bind function arguments to the model.
-// "functions.parameters" must be formatted as JSON Schema.
-// We translate the above Zod schema into JSON schema using the "zodToJsonSchema" package.
-// Omit "function_call" if you want the model to choose a function to call.
 const model = new ChatOpenAI({
   modelName: "gpt-4",
 }).bind({
-  functions: [
-    {
-      name: "extractor",
-      description: "Extracts fields from the input.",
-      parameters: zodToJsonSchema(extractionFunctionZodSchema),
-    },
-  ],
+  functions: [extractionFunctionSchema],
   function_call: { name: "extractor" },
 });
 
 const result = await model.invoke([new HumanMessage("What a beautiful day!")]);
 
 console.log(result);
-
 /*
-  AIMessage {
-    content: '',
-    name: undefined,
-    additional_kwargs: {
-      function_call: {
-        name: 'extractor',
-        arguments: '{\n' +
-          '  "tone": "positive",\n' +
-          '  "entity": "day",\n' +
-          '  "word_count": 4,\n' +
-          '  "chat_response": "It certainly is a gorgeous day!",\n' +
-          '  "final_punctuation": "!"\n' +
-          '}'
-      }
+AIMessage {
+  lc_serializable: true,
+  lc_kwargs: { content: '', additional_kwargs: { function_call: [Object] } },
+  lc_namespace: [ 'langchain', 'schema' ],
+  content: '',
+  name: undefined,
+  additional_kwargs: {
+    function_call: {
+      name: 'extractor',
+      arguments: '{\n' +
+        '"tone": "positive",\n' +
+        '"entity": "day",\n' +
+        '"word_count": 4,\n' +
+        `"chat_response": "I'm glad you're enjoying the day!",\n` +
+        '"final_punctuation": "!"\n' +
+        '}'
     }
   }
+}
 */