Skip to content

Latest commit

 

History

History
223 lines (130 loc) · 7.28 KB

keras_evaluation.md

File metadata and controls

223 lines (130 loc) · 7.28 KB
Raw

Project Name: Keras

Evaluating Person or Team:

Mark @markflaherty

Project Data

  1. Project description:
    Keras is a high-level neural networks API, written in Python and capable of running on top of TensorFlow, CNTK, or Theano.

  1. Project website/homepage: https://keras.io/

  2. Project repository:

    https://github.com/keras-team/keras

License

  1. What is the project's license?
    MIT

Code Base

  1. What is the primary programming language in the project? Python

  2. What is the development environment?

    It does not need a certain enviroment to run, however it is dependent upon TensorFlow

  3. Are there instructions for how to download, build, and install? How easy is it to find them? Do they seem easy (relatively speaking) to follow?

    There are well laid out instructions to download. I would say that they are not necessarily basic instructions, however, the help is there.

  4. Does the project depend on external additional software modules such as database, graphics, web development, or other libraries? If so, are there clear instructions on how to install those?

    Yes, and there is instructions to get all of them. The library is dependent upon TensorFlow, Theano, CNTK.

  5. Is the code easy to understand? Browse some source code files and make a judgment based on your random sample.

    Yes, if someone is familiar with python, the files aree very easy to file.

  6. Is this a big project? If you can, find out about how many lines of code are in it, perhaps on OpenHub.

    49.6k lines of code, so a decent sized project.

  7. Does the repository have tests? If so, are the code contributors expected to write tests for newly added code?

    There are tests but there is not any guidelines for whether or not you should add tests to your updated code. From looking at the tests, it seems that you can use their tests, as though they are quite extensive.

Code and Design Documentation

  1. Is there clear documentation in the code itself?

    On some of the files their is extensive documentation, but on others there is not as much.

  2. Is there documentation about the design?

    Very sparse documentation on design, it is limited to specific files.

Activity Level

  1. How many commits have been made in the past week?

    0

  2. When was the most recent commit?

    November 3rd

  3. How many issues are currently open?

    2,849

  4. How long do issues stay open?

    I would say on average 3-5 days.

  5. Read the conversations from some open and some closed issues. Is there active discussion on the issues?
    Almost all issues that have been closed have an active discussion in them

  6. Are issues tagged as easy, hard, for beginners, etc.?

    The closest it comes is the "Good First Issue Tag"

  7. How many issues were closed in the past six months?
    108

  8. Is there information about how many people are maintaining the project?

    Yes, under the network tab of insights it shows that Keras is mostly developed by their team and 7 other developers

  9. How many contributors has the project had in the past six months?

    There hasn't been activity outside of the team, or the 7 other developers in the past 6 months

  10. How many open pull requests are there?

    71

  11. Do pull requests remain un-answered for a long time?

    Usually closed within a week.

  12. Read the conversations from some open and some closed pull requests. Is there active discussion on the pull requests?

    Yes their is a pretty length conversation on the pull requests I saw

  13. How many pull requests were opened within the past six months?

    63

  14. When was the last pull request merged?

    November 6th

Welcomeness and Community

  1. Is there a CONTRIBUTING document? If so, how easy to read and understand is it? Look through it and see if it is clear and thorough.

    There is a pretty thorough contributing document, laying out what they want from people to contribute and how to report bugs in their fashion.

  2. Is there a CODE OF CONDUCT document? Does it have consequences for acts that violate it?

    They do not have a code of conduct document, github allows you to propose one to them

  3. Do the maintainers respond helpfully to questions in issues? Are responses generally constructive? Read the issue conversations.

    Not really, its either a follow up question or a close. Otherwise, the conversation is quite dry and it seems like they all know what they are doing.

  4. Are people friendly in the issues, discussion forum, and chat?

    It seems like the are friendly.

  5. Do maintainers thank people for their contributions?

    No.

Development Environment Installation

Install the development environment for the project on your system. Describe the process that you needed to follow:

  1. how involved was the process?

    I have used Keras before, so I have already installed it. The process did not take long as though the instructions are pretty straight forward.

  2. how long it take you?

    10 minutes

  3. did you need to install additional packages or libraries?

    Yes, TensorFlow is a must

  4. were you able to build the code following the instructions?

    Yes.

  5. did you need to look for additional help in installing the environment?

    No

  6. any other comments?

Summary

  1. Do you think this is a project to which it would be possible to contribute in the course of a few weeks before the end of this semester?

    I think I could contribute to this project, especially because they do not have a code of conduct, and that is something important to have when trying to continue open source development. Likewise, it is something that I am very interested it whether or not I get to touch the code.

  2. Would you be interested in contributing to this particular project?

    Yes, like I said above, I have been using this project in my own time. I really like the features that they support and their website is very conductive to learning about ML. Moreover, I think being able to work on this project would be something very cool for me to do.