Forward propagation = Prediction
Back propagation = Measuring i.e. learning
jupyter - Online IDE
kaggle - datasets
Wrappers to simplify using AI models to do tasks. Basically write and orchestrate prompts indirectly.
OpenAI API is quite easy to use directly and gives you the most control, I don’t see much benefit of a wrapper langChain here tbh.
Example: Langchain, Haystack, Microsoft Guidance