Lesson proposal: Creating Deep Neural Networks for Regression Analysis Proposal (PH/JISC/TNA)

[tiagosousagarcia]

The Programming Historian has received the following proposal for a lesson on 'Creating Deep Neural Networks for Regression Analysis Proposal' by @nabsiddiqui. The proposed learning outcomes of the lesson are:

• Investigate how machine learning can provide humanities scholars new tools for image analysis.
• Generate a deep neural network for regression analysis that provides a range of outputs for different paintings.
• Assess the role that different layers, architectures, and tasks of a neural network function.

In order to promote speedy publication of this important topic, we have agreed to a submission date of no later than 24/01/2022. The author(s) agree to contact the editor in advance if they need to revise the deadline.

If the lesson is not submitted by 24/01/2022, the editor will attempt to contact the author(s). If they do not receive an update, this ticket will be closed. The ticket can be reopened at a future date at the request of the author(s).

The main editorial contact for this lesson is @tiagosousagarcia.

Our dedicated Ombudsperson is (Ian Milligan - http://programminghistorian.org/en/project-team). Please feel free to contact him at any time if you have concerns that you would like addressed by an impartial observer. Contacting the ombudsperson will have no impact on the outcome of any peer review.

[scottkleinman]

Nope. I think @anisa-hawes will take it from here.

[drjwbaker]

@anisa-hawes: do you need some help getting the copyediting done here? (I know you have a lot on!)

[anisa-hawes]

Happy New Year @drjwbaker! Apologies for the delay: these edits are ready for @nabsiddiqui and @scottkleinman's review.

[nabsiddiqui]

Hey @drjwbaker. Just FYI, I plan to get these done sometime early next week.

[drjwbaker]

@nabsiddiqui Nearly there!

[nabsiddiqui]

@drjwbaker yes, I just need to confirm the copy edits. I have to look up how to do them with a code review as I have never done them that way before.

[nabsiddiqui]

@drjwbaker @anisa-hawes

Sorry for the delay on this. I looked at it multiple times last week, and it seems that I had just over analyzed how difficult it would be to actually get this done through a code review. It didn't take long at all. Let me know what else is needed on my end.

[hawc2]

@drjwbaker @anisa-hawes just checking in. it sounds like nabeel's edits are done. can this lesson be copyedited soon? let me know when you'd like me to review it

[drjwbaker]

Amazing. I'll leave to @anisa-hawes to confirm the copyediting process (remembering, Anisa, that we have budget for this).

[anisa-hawes]

Hello @hawc2,

Apologies, as I failed to post this summary of my actions! This lesson has already been copyedited and it is ready for your final review.

• Preview: http://programminghistorian.github.io/ph-submissions/en/drafts/originals/image-classification-neural-networks
• .md file: /drafts/originals/lesson-file-name-here.md
• images: /images/image-classification-neural-networks
• assets: /assets/image-classification-neural-networks
• original avatar: /gallery/originals/image-classification-neural-networks.png
• gallery avatar: /gallery/lesson-file-name-here.png

Sustainability + accessibility actions status :

• [x] Copyediting
• [x] Typesetting
• [x] Addition of Perma.cc links
• [x] Addition of alt-text for all figures
• [x] Receipt of authorial copyright agreement

Our authorial copyright declaration form is an opportunity to acknowledge your copyright and formally grant us permission to publish. @nabsiddiqui could you download this, complete the details, and email it to me (admin [@] programminghistorian.org)?

Next steps @scottkleinman :

• [x] Select + upload an image for the lesson (incl. avatar_alt:)
• [ ] Prepare x2 posts for our Twitter/Mastodon Bot
• [x] As @nabsiddiqui is already included in ph_authors.yml (and the information is up-to-date), we do not need to make any additions to that file

[nabsiddiqui]

Just sent the form.

[scottkleinman]

I see that I am tagged to prepare posts for Twitter/Mastodon. @anisa-hawes, is there a procedure for this described somewhere? I'm not sure what steps to take.

[anisa-hawes]

Hello @scottkleinman

Sorry not to have clarified this. We are in flux at the moment, but will soon be re-establishing our Twitter/Mastodon bots.

We ask for 2 x posts about each lesson (what can you learn from it?who contributed to it?), which will be posted by our Bots within a given week. Each post will also contain the DOI link to the lesson.

The original (set up for Twitter handles) Google Sheet is here. Let me know if you have any difficulties accessing it -- as a member of our Google Group you should be able to edit.

Message one (Column A) + DOI is posted early in a week. Message two (Column B) + DOI is a follow-up to be posted later the same week.

These 2 posts (unlike an "announcement" post upon publication) are to publicise the lesson in the future so should not mention that the lesson is "new". (Our bots select one lesson each week at random, so it could be a while until the Bots select this lesson).

If you know/can find Mastodon account handles for the lesson contributors, please email me. That would be a great help as I can prepare variations of each post for both platforms.

(I am in the process of figuring out how to make all that ^ simple).

[scottkleinman]

OK, I'll hold off for the moment, now that I have determined that I can access the Google Sheet. Let me know when the DOI becomes available.

[anisa-hawes]

Thank you, @scottkleinman. The DOI will not be activated until publication. So the task here is just to prepare the wording for those two posts, so we can slot them into the Google Sheet when the time comes.

[hawc2]

This looks good. One thought @scottkleinman @nabsiddiqui. The Table of Contents is a little different from most lessons, which usually have a more hierarchal structure.

I wonder if this could be broken into a couple Parts, like "Set Up," "Understanding Neural Networks," and "Creating a Model" or something like that? Then the individual sections could be subcategories of those two-three parts of the lesson? Anyway, this should be an easy change if you agree with it.

I would like to fasttrack this to publication, but we currently have a Jekyll build error (see #2879) for another lesson I'm trying to publish, and we'll need to figure out first what's causing it.

[scottkleinman]

Here's one possible way to organise the TOC:

• Introduction
  • Audience and Requirements
• Setup and Dataset
• Understanding Neural Networks
  • A Basic Neural Network
  • Convolutional Neural Networks
  • Transfer Learning and Convolutional Neural Networks
• Creating Your Own Model (right before paragraph 39)
  • Export the Model
  • Import the Model with ml5.js
• Conclusion

What say you, @nabsiddiqui?

[nabsiddiqui]

@scottkleinman works for me!

[scottkleinman]

@hawc2, I have added the hierarchy suggested above. It should be propagated in the next successful build.

[nabsiddiqui]

@anisa-hawes Is this set to publish?

[anisa-hawes]

Thank you @scottkleinman and @nabsiddiqui! Yes!

This is ready for @hawc2 to prepare for publication.

[hawc2]

Thanks all! This is next on my docket

[hawc2]

@scottkleinman @nabsiddiqui I read through this and think it's pretty much ready to go. I made some minor line edits. I'm really impressed with how well this lesson explains such a difficult subject, so well done! One thought for revision - can you change the last section so it's broken up into Conclusion and References? Instead of that list of references with little dashes before it, let's just turn that into an official References section. Once you make that change, I'll start drafting this for publication

[scottkleinman]

@hawc2 Done!

[hawc2]

hey all, this lesson is now published and viewable online!: https://programminghistorian.org/en/lessons/image-classification-neural-networks

Congrats @nabsiddiqui! And big thanks to @zentralwerkstatt and @melvinwevers for reviewing this lesson, as well as to @scottkleinman for editing it!

[nabsiddiqui]

Thank you everyone for all your work on this. I really appreciate it.