Just a stranger trying things.

  • 1 Post
  • 13 Comments
Joined 1 year ago
cake
Cake day: July 16th, 2023

help-circle
  • In the deep learning community, I know of someone using parquet for the dataset and annotations. It allows you to select which data you want to retrieve from the dataset and stream only those, and nothing else. It is a rather effective method for that if you have many different annotations for different use cases and want to be able to select only the ones you need for your application.












  • Im not sure what the issue is. I have listened to Chris talk about the development of Mojo and while I have not come across any confirmation for open sourcing all aspects of Mojo yet (which would be desirable), the fact that it is being developed with involvement from the community, in a closed beta is understandable and also a good idea.

    He explained it as this: during the closed beta, he didn’t want people to start sharing temporary quirks and bugs or features while they could still drastically change. Someone complained at some point that the python function “open” was not recognized. That’s probably exactly what they are set up to solve and we would probably here many more “issues” from people if it were all happen in the open. People would just see it’s not working perfectly as if it were released and would dismiss it and it could mean the end of Mojo.

    Instead, they selected individuals who cared, who wanted to be involved and improve it and give feedback. This is a normal development process. It is logical to me to launch it once the language is mature and most issues are ironed out and that the API, language and features are more stable.

    Edit: what are people who downvote disagreeing about?