AI Applications Made Easy: Dive into LangChain & GPT

Explore the two model types in long chain models—large language models (LMS) and chat models—and understand their features and use cases.

Explore large language models that turn data into reports and offer medical guidance, integrated with LangChain and providers like OpenAI, Cohere, and HuggingFace.

Explore hands-on setup with LangChain and OpenAI in a Jupyter notebook, initialize the OpenAI LM, run simple prompts, and execute batch queries with the generate method to handle multiple prompts.

Explore prompt templates in LangChain to craft dynamic, context-aware prompts. Learn how predefined recipes with instructions, examples, and content create model-agnostic prompt values.

Follow a live demo to build a simple prompt template in LangChain, import modules in a Jupyter notebook, and generate a dynamic, context aware prompt ready for language models.

Explore fused prompt templates and few-shot prompt templates to blend templates with few-shot learning, focusing on self-park with search and formatting examples with a prompt template object for Lang Chain.

Explore few-shot learning with OpenAI and LangChain, using example selectors and chroma for semantic similarity search to craft context-rich prompts and accurate responses.

Build a joke generator with OpenAI and Langton's output parser by importing libraries, authenticating with an API key, crafting a prompt template, and parsing model output with a pedantic parser.

Import the LangChain and OpenAI libraries in a notebook, authenticate with an API key, set up a Pydantic model and prompt template, and build a pedantic output parser for jokes.

A chain acts as your personal query guide, taking your question, packaging it neatly, and sending it to the language model to return a neatly formatted answer.

Harness OpenAI's GPT and the Langson library to compose lyrics on a topic, then verify their appropriateness via a sequential chain.

Explore how memory and context power conversational systems by remembering previous conversations to drive future responses in large language models. See why systems that forget disrupt dialogue, and how memory underpins coherent interactions in AI applications.

Store and retrieve past interactions with Lang Tune's memory module, which can be standalone or embedded in a Lang chain, reads before logic and writes after, including chat message history.

Explore memory integrated chat bots using the Langson library and OpenAI, as memory and conversation chains create context-aware interactions with the GPT model.

Discover how indexes fetch up-to-date information for language models by retrieving relevant documents via LangChain, using document loaders and text splitters to boost retrieval performance.

Explore document loaders that translate raw, unstructured data into clean documents with metadata for LangChain. Use the load method for one-shot imports or lazy load for large datasets.

Import the text file loader from the Lang change document loaders package, load hello.txt, and print the long chain Langton document object with text and metadata.

Explore text splitters in lang chain to break long documents into chunks that preserve context. Compare splitter strategies and measurements, with the recursive character splitter as the default.

Import the recursive character text splitter from LangChain, initialize it with chunk size and overlap, and use create_documents to split a 300-page paper into bite-sized, context-preserving chunks for analysis.

Explore embeddings that convert text and categorical data into numerical representations, learned from data to map items into a lower dimensional space and reveal meaningful relationships.

Explore vector stores as a Google-like index for your document collection, enabling rapid retrieval of the most relevant papers to a query. Apply to climate change research and customer support.

Set up a chroma vector store in a Jupyter notebook, embed and store document chunks, then search papers on carbon pricing to demonstrate practical retrieval.

Discover how LangChain agents use a language model to make dynamic decisions, employ a toolkit of tools, and balance real-time actions with predefined workflows.

Explore AI agents powered by LangChain and GPT that use tools like Wikipedia and calculator, authenticate with API keys, and convert queries into actions and answers.

Explore a Jupyter walkthrough of text retrieval and question answering with LangChain and GPT, from data loading and embeddings to semantic search with a multi query retriever and source citations.

Explore a business chatbot that delivers instant, accurate responses and detailed product explanations to boost user experience. Maintain chat histories for seamless, referable conversations across complex interactions.

Explore a Streamlit chat bot code walkthrough built with OpenAI's GPT and the Langson library, demonstrating memory, a prompt system, real-time interaction, and key handling via constants file.