¶ Guides
These tutorials in a lesson-like form are designed to show you how to take full advantage of our system.
Looking to set up your first assistant, use some advanced features, or troubleshoot any issues? You will find everything here.
Start exploring and take your skills to the next level!
¶ Text Guides
First thing you may want to do is take one of the ready-made Dream distributions, run it and have a chat. To do that, take a look at the beginner’s tutorial on how to get started with Dream.
Once you are done, you can move on to the tutorials:
-
Create a new assistant by modifying the existing Dream distributions. Discover how to create a generative bot with a predefined persona, using the existing Dream Persona distribution that utilizes OpenAI ChatGPT.
-
Create a new distribution utilizing the existing Dream components. Learn how to create a simple and light-weight bot that specializes in discussing movies and answering factoid questions, utilizing the existing Dream components.
-
Creating your own Generative AI Assistant — if you are mainly interested in large language models and their capabilities, this tutorial is for you. You will learn how to integrate different models into Dream and try them out using your own prompts.
-
Create a distribution with reasoning and API integration utilizing OpenAI ChatGPT to think of actions required to handle user requests and to choose the suitable API to use.
-
Create a distribution for question answering based on the given documents. Document-based LLM QA is a distribution of Dream designed to answer questions about the content of one or several documents supplied by the user. This distribution uses TF-IDF vectorization and cosine similarity to detect parts of documents most relevant to the user’s question. N most relevant parts, alongside with an instruction and the dialog context are then passed on to ChatGPT as a prompt.
-
Adding your custom DFF skill to an existing assistant — skills are key building blocks that define the responses your system gives. Read this tutorial to understand how to create your own skill using DeepPavlov Dialog Flow Framework (DFF) — our framework for creating and managing custom scenario-driven conversational AI skills. DFF is the most useful for rule-based assistants.
¶ Additional Tutorials
-
Prompting for response generation helps to discover more about different LLMs and choosing between them.
-
Extracting dialogs from a database — if you want to get the dialogs that you or your users had with your system, take a look at this tutorial.
-
Utilizing separate components in Dream - if you want to utilize particular component from the existing distribution.
-
Special Responses to Intents and FAQ — in some cases, you may want to give strictly pre-defined responses to some user utterances. To learn how to do that, check out this tutorial.
-
How to utilize Proxy containers by DeepPavlov Dream helps to understand what proxy is and hot to use it to save you own computational resources.
¶ Video Guides
We are actively working on adding new video guides. Please, find below a bit old but still actual video tutorials.
If you have already mastered all the other tutorials and want to learn more, take a look at these videos:
Couldn’t find what you were looking for? Contact us!