1. What does an Executor do?

  • An Executor is the tool you use to execute your Flow from the CLI: execute jina_app.py
  • An Executor is a processing pipeline for indexing or querying a dataset
  • An Executor performs a single task on a Document or DocumentArray, like segmenting or encoding it.

2. What does a Flow do?

  • The Flow ties Executors together into a processing pipeline to perform a bigger task, like indexing or querying a dataset.
  • The Flow is a graphical interface that lets users see how their Documents are flowing into the processing pipeline.
  • The Flow is short for "fast, low-resource" and is a special kind of Executor for low-powered machines.
  • The Flow helps you visualize the input and the output flow of data.

3. What's the recommended way to share Executors with a colleague?

  • Send them a link to the repo
  • Dockerize your Executor and push directly to Docker Hub
  • Push your Executor to Pypi and ask them to install via pip
  • Push your Executor to Jina Hub and share

4. Given a Document doc, what does doc.content refer to?

  • doc.tensor
  • doc.blob
  • doc.text
  • Any of the above as long as the field is not empty

5. What is the best way to embed Documents in a DocumentArray

  • loop over Documents and compute embeddings
  • Iterate over batches of Documents and compute embeddings in batches
  • Use method DocumentArray.embed
  • Create a new DocumentArray with computed embeddings

6. What formats are supported for creating Executors?

  • YAML
  • JSON
  • Python
  • YAML and Python

7. What function decorator is used to invoke an Executor?

  • @requests
  • @start
  • @invoke
  • @jina_invoke

8. What are the different ways to send request data in a Flow?

  • Document, DocumentArray, and None
  • An iterator of Document
  • A generator of Document
  • All of them

9. How can I store a DocumentArray in a database?

  • Use a different DocumentArray store
  • Serialize the DocumentArray and store it in a database
  • Create a relational database table for Document and use an ORM to persist Document to the table

10. Which communication protocol(s) does the Flow support?

  • HTTP, ZMQ and Sockets
  • GRPC, Websockets and HTTP
  • GRPC, HTTP and ZMQ
  • Websockets, GRPC and FTP
Quick access to:

Beginner Section


Intermediate Section


Advanced Section