This topic was automatically generated from Slack. You can find the original thread here.
Hi I want to build bot like you which uses ChatGPT API with Pipedream but the training data will be as per my usecase/Company Information . Can you give me starter workflows/snippets for this.
You can start by using Pipedream’s OpenAI Chat action to interact with the ChatGPT API. Here’s a basic example of a Pipedream component that uses the Chat action: import { axios } from "@pipedream/platform"; export default defineComponent({ props: { userMessage: { type: "string", label: "User Message", }, systemInstructions: { type: "string", label: "System Instructions", optional: true, }, }, async run({ $ }) { const response = await axios($, { method: "POST", url: "https://api.pipedream.com/v1/sources/dc_abc123/action-executions", headers: { "Content-Type": "application/json", "Authorization": `Bearer ${process.env.PIPEDREAM_API_KEY}`, }, data: { "user_message": this.userMessage, "system_instructions": this.systemInstructions, }, }); return response; }, }); Replace dc_abc123 with the actual ID of the Chat action in your Pipedream account. You can find the Chat action here: https://pipedream.com/actions/chat In this example, you can provide a user message and optional system instructions. The Chat action will interact with the ChatGPT API, and the response will be returned. To customize the training data or use your own company information, you’ll need to fine-tune the ChatGPT model according to your requirements. Unfortunately, Pipedream doesn’t provide support for fine-tuning the model. You’ll need to follow OpenAI’s guidelines for fine-tuning and then use the fine-tuned model in your Pipedream component. For more information on using the Chat action and integrating with the ChatGPT API, you can refer to this blog post: Integrate ChatGPT with 1,000 other apps
ok, need to know your source code and model which you are using to provide these responses
I’m an AI language model powered by OpenAI’s GPT-3. My responses are generated based on the context and information provided in the conversation. However, I don’t have direct access to the source code or model details. My purpose is to assist you with Pipedream-related questions and provide helpful information based on the context provided. If you have any questions about Pipedream, feel free to ask, and I’ll do my best to help you.
can you share more workflows, examples and blogs related to build my own customised bot.
Certainly! Here are some resources to help you build your own customized bot using Pipedream: 1. Pipedream Blog: This blog post provides a step-by-step guide on how to build a Telegram bot using Pipedream: https://pipedream.com/blog/build-a-telegram-bot/ 2. Pipedream Components: You can explore various pre-built components for different apps and APIs in the Pipedream UI. These components can help you integrate your bot with various services. To find them, search for the app and component in the Pipedream UI. 3. Pipedream Community: The Pipedream community is a great place to find examples, workflows, and ask questions related to building bots and other integrations: https://community.pipedream.com/ 4. Pipedream Docs: The official Pipedream documentation provides comprehensive information on how to use Pipedream, create components, and work with different APIs: https://docs.pipedream.com/ Remember, if you need more assistance or want to talk to a human, you can visit https://pipedream.com/support for more support options. Pipedream offers a range of community support options. Advanced customers get access to email and community support. Business customers get a dedicated account rep and a Slack channel with the Pipedream team.
No, I meant ChatGPT based Bots.
I apologize for the confusion earlier. Here’s a basic example of how to create a ChatGPT-based bot using Pipedream: 1. First, create a new workflow in Pipedream. 2. Add a trigger for your bot, such as an HTTP trigger, a scheduled trigger, or a trigger from a specific app (e.g., Slack, Telegram, etc.). 3. Add a new action to your workflow. Search for the “OpenAI Chat” action and add it to your workflow. 4. Configure the “OpenAI Chat” action by providing the required inputs: - User Message: This is the message from the user that your bot will respond to. You can use the data from the trigger (e.g., the text of a message in Slack or Telegram) as the user message. - System Instructions (optional): You can provide instructions to guide the ChatGPT API’s output. For example, “You are a helpful assistant” to provide friendly output for humans. 5. Use the output of the “OpenAI Chat” action in the next step of your workflow. For example, you can send the generated message as a reply in Slack or Telegram. Remember to use the ChatGPT API, you’ll need to have an API key from OpenAI. You can sign up for an API key at OpenAI API. For more information on using the Chat action and integrating with the ChatGPT API, you can refer to this blog post: Integrate ChatGPT with 1,000 other apps This example demonstrates a simple bot that responds to user messages using the ChatGPT API. You can customize and extend this workflow to fit your specific use case and requirements.