Merge pull request #1080 from Agenta-AI/feat/templates

Add async version of baby name generator app

examples/baby_name_generator/app.py

@@ -1,8 +1,8 @@
+from agenta import FloatParam, TextParam
 import agenta as ag
 from openai import OpenAI
 
 client = OpenAI()
-from agenta import FloatParam, TextParam
 
 default_prompt = (
     "Give me 10 names for a baby from this country {country} with gender {gender}!!!!"

examples/baby_name_generator/app_async.py

@@ -0,0 +1,34 @@
+from agenta import FloatParam, TextParam
+import agenta as ag
+from openai import AsyncOpenAI
+
+client = AsyncOpenAI()
+
+default_prompt = (
+    "Give me 10 names for a baby from this country {country} with gender {gender}!!!!"
+)
+
+ag.init()
+ag.config.default(
+    temperature=FloatParam(0.2), prompt_template=TextParam(default_prompt)
+)
+
+
+@ag.entrypoint
+async def generate(country: str, gender: str) -> str:
+    """
+    Generate a baby name based on the given country and gender.
+
+    Args:
+        country (str): The country to generate the name from.
+        gender (str): The gender of the baby.
+
+    Returns:
+        str: The generated baby name.
+    """
+    prompt = ag.config.prompt_template.format(country=country, gender=gender)
+
+    chat_completion = await client.chat.completions.create(
+        model="gpt-3.5-turbo", messages=[{"role": "user", "content": prompt}]
+    )
+    return chat_completion.choices[0].message.content

examples/job_info_extractor/README.md

@@ -1,9 +1,49 @@
-# Using this template
+# Extraction using OpenAI Functions and Langchain"
 
-Please make sure to create a `.env` file with your OpenAI API key before running the app.
-OPENAI_API_KEY=sk-xxxxxxx
 
-You can find your keys here:
-https://platform.openai.com/account/api-keys
+This templates is designed to extracts job information (company name, job 
+title, salary range) from a job description. It uses OpenAI Functions and 
+Langchain. It runs with agenta. 
+[Agenta](https://github.com/agenta-ai/agenta) is an open-source LLMOps 
+platform that allows you to 1) quickly experiment and compare 
+configuration for LLM apps 2) evaluate prompts and workflows 3) deploy 
+applications easily. 
+
+## How to use
+### 0. Prerequisites
+-  Install the agenta CLI
+```bash
+pip install agenta-cli
+```
+- Either create an account in [agenta cloud](https://cloud.agenta.ai/) or 
+[self-host agenta](/self-host/host-locally)
+
+### 1. Clone the repository
+
+```bash
+git clone https://github.com/Agenta-AI/job_extractor_template
+```
+
+### 2. Initialize the project
+
+```bash
+agenta init
+```
+
+### 3. Setup your openAI API key
+Create a .env file by copying the .env.example file and add your openAI 
+API key to it.
+```bash
+OPENAI_API_KEY=sk-xxxxxxxxxxxxxxxxxxxxxxxx
+```
+
+### 4. Deploy the application to agenta
+
+```bash
+agenta variant serve app.py
+```
+
+### 5. Experiment with the prompts in a playground and evaluate different variants in agenta
+
+https://github.com/Agenta-AI/job_extractor_template/assets/4510758/30271188-8d46-4d02-8207-ddb60ad0e284
 
-Go back to the [Getting started tutorial](https://docs.agenta.ai/getting-started) to continue

examples/job_info_extractor/app.py

@@ -11,11 +11,17 @@
 
 from pydantic import BaseModel, Field
 
-default_prompt = "What is a good name for a company that makes {product}?"
+CHAT_LLM_GPT = [
+    "gpt-3.5-turbo-16k-0613",
+    "gpt-3.5-turbo-16k",
+    "gpt-3.5-turbo-0613",
+    "gpt-3.5-turbo-0301",
+    "gpt-3.5-turbo",
+    "gpt-4",
+]
 
 ag.init()
 ag.config.default(
-    prompt_template=ag.TextParam(default_prompt),
     system_message=ag.TextParam(
         "You are a world class algorithm for extracting information in structured formats."
     ),
@@ -26,10 +32,11 @@
     company_desc_message=ag.TextParam("The name of the company"),
     position_desc_message=ag.TextParam("The name of the position"),
     salary_range_desc_message=ag.TextParam("The salary range of the position"),
-    temperature=ag.FloatParam(0.5),
-    top_p=ag.FloatParam(1.0),
+    temperature=ag.FloatParam(0.9),
+    top_p=ag.FloatParam(0.9),
     presence_penalty=ag.FloatParam(0.0),
     frequency_penalty=ag.FloatParam(0.0),
+    model=ag.MultipleChoiceParam("gpt-3.5-turbo-0613", CHAT_LLM_GPT),
 )
 
 
@@ -50,7 +57,7 @@ def generate(
 ) -> str:
     """Extract information from a job description"""
     llm = ChatOpenAI(
-        model="gpt-3.5-turbo-0613",
+        model=ag.config.model,
         temperature=ag.config.temperature,
         top_p=ag.config.top_p,
         presence_penalty=ag.config.presence_penalty,