feat: add gpt3 function example

Author: datitran

README.md

@@ -1,8 +1,8 @@
-# 🧶 Priceloop NoCode Python External Functions Templates
+# 🧶 Priceloop NoCode Python Custom Functions Templates
 
 You can create your own functions and then call them inside our [NoCode table](https://priceloop.ai/nocode), just like a built-in function. Here are some easy examples that you can use to write your own functions. Some of them are also deployed on our platform.
 
-https://user-images.githubusercontent.com/3866530/200024328-ea5cad88-c930-4544-a9d4-9838437978f0.mp4
+<https://user-images.githubusercontent.com/3866530/200024328-ea5cad88-c930-4544-a9d4-9838437978f0.mp4>
 
 ### Getting Started
 
@@ -13,7 +13,7 @@ conda create -n my_example python=3.8
 source activate my_example
 ```
 
-The easiest way to deploy our external functions is via our cli:
+The easiest way to deploy our custom functions is via our cli:
 
 ```bash
 npm install -g @priceloop/cli
@@ -36,7 +36,7 @@ After these steps, your new function will be available inside the specified work
 
 Go to the app and type a new formula: `\my_example('World')`. It should return “Hello World”.
 
-More information on how to deploy it to our platform please also go to our [documentation](https://priceloopai.notion.site/External-Functions-f78d153ab7b94a5f8a2f2cc5baa5e9d3).
+More information on how to deploy it to our platform please also go to our [documentation](https://docs.priceloop.ai/custom_functions).
 
 ### Testing
 

gpt3/README.md

@@ -0,0 +1,11 @@
+# GPT3 Example
+
+## Testing locally
+
+1. Go to <https://openai.com/api/> and register an account
+2. Get the API key and replace the `<openai-key>` in the `event.json` with your key
+3. Run this to test:
+
+```
+python-lambda-local function.py event.json -t 900
+```

gpt3/event.json

@@ -0,0 +1,3 @@
+{
+    "BatchedArgs": [["Extract name from email: [email protected]", "<openai-key>"], ["Extract name from email: [email protected]", "<openai-key>"]]
+}

gpt3/function.py

@@ -0,0 +1,23 @@
+import json
+import requests
+
+def return_gpt3_results(input, api_key):
+    headers = {
+        "Authorization": "Bearer {}".format(api_key),
+    }
+
+    json_data = {
+        "model": "text-davinci-002",
+        "prompt": input,
+        "temperature": 0.8,
+        "max_tokens": 60,
+        "top_p": 1,
+        "frequency_penalty": 0,
+        "presence_penalty": 0,
+    }
+    response = requests.post("https://api.openai.com/v1/completions", headers=headers, json=json_data)
+    return json.loads(response.content.decode("utf-8"))["choices"][0]["text"]
+
+def handler(event, context):
+    api_key = list(map(lambda row: row[1], event["BatchedArgs"]))[0]
+    return list(map(lambda row: return_gpt3_results(row[0], api_key), event["BatchedArgs"]))

gpt3/requirements.txt

@@ -0,0 +1 @@
+requests