How do we all keep going with deep learning?

Use this forum thread to discuss how we can keep going with deep learning. Making commitments in writing can help you follow through, and this thread is a great place to do that. We will keep the wiki, forums, and slack channels going.

Some questions:

  • What next steps do you plan to take? This could include reviewing material we covered (particularly if you were strapped for time during this course), exploring new data sets, starting new projects at work, or continuing with projects (such as Cats vs Dogs or the Fisheries Competition) that you’ve already started.
  • What do you need from us to continue with your learning?
  • What challenges do you anticipate facing in the coming weeks?
  • Most importantly, how can we keep this community going and make the most of it?
2 Likes
  • What next steps do you plan to take?
    I’d like to go back over cats vs dogs and the fisheries competition to implement all of the things we’ve learnt, and explore the impact the different tools have.
    I’d like to start a project of my own around audio processing.
    A stretch goal would be to enter a currently running kaggle competition and get in the top 3rd.

  • What do you need from us to continue with your learning?
    Support on the forums will be highly appreciated!

  • What challenges do you anticipate facing in the coming weeks?
    Staying motivated with all the Christmas/NY distractions.

  • Most importantly, how can we keep this community going and make the most of it?
    One of the cool things about this course has been the cutting edge updates. A few of the things we’ve talked about have come out in the last 6 months, other things were implemented by Jeremy during the prior week. Keeping the group in the loop on new developments, and maybe even encouraging people to help contribute new techniques to Keras or other tools.

1 Like

I think I’d like a helpdesk channel on the slack for any of the problems that we looked at but still run on our own. Even if the help was here’s the link from the forum might help.

or some group projects that people would like help on feels like it should work.

What challenges do you anticipate facing in the coming weeks?
Finding problems or topics that a beginner practiioner can help with.

Most importantly, how can we keep this community going and make the most of it?
holding a space open for the community to keep meeting each other to share knowledge, projects, jobs.

1 Like

This has been a great dive into deep learning for me. While I was familiar with some of the concepts of deep learning, this class really demystified things for me. I have two next steps on my list: 1) continue the practice by participating in Kaggle competitions, 2) study the underlying theory to get a better intuitive understanding. For this second part I’m considering the Coursera class by Hinton, but also looking around at books and online tutorials.

This has been a very encouraging start to deep learning - so thank you all.

What next steps do you plan to take?

  • I would be going over cats/dogs, statefarm and fisheries

What do you need from us to continue with your learning?

  • i would love to have access to the notebooks shown during class as well as the slack channels

What challenges do you anticipate facing in the coming weeks?

  • with the holidays coming, i am planing to spend my time going over the course materials. the challenge i see is that i might not be able to reach anyone for help during the holidays and get stuck.

Most importantly, how can we keep this community going and make the most of it?

Additionally, during this part of the year i was very busy at work and didnt get a chance to spend the time to network and get to know more people in the class. I would love to know if there will be folks interested in meetups/work sessions during the holidays (Dec 23 - Jan 1).

3 Likes

I interviewed Pedro Domingos for my podcast recently, he noted that the only way to learn machine learning is to do it. Just like learning to drive, you won’t learn much from a book unless you’re also practising. However, I’m wary of disappearing down a rabbit hole without first choosing the right tool for the job so I’m looking forward to hearing what everyone has to say at the AIMed conference.

I could use some help transitioning from the problems presented in the classes to ones that I need to solve in my professional life. Or perhaps more importantly, I need help finding the problems that I should be solving in my professional life.

Finding time to play with something that may yield dividends in the future is always hard when you work freelance and bill by the hour.

My suggestion is to keep in the intakes up at a healthy rate. No doubt that will happen once the MOOC is online, you may want to consider answering questions found on this forum in public on the blog so that people start finding you with organic search.

You’re all doing fabulous work, thank you!

2 Likes

Could you help me understand what you’re looking for here? In what way do these forums not provide what you need for such a ‘help desk’. (I’ve found that Slack tends to result in the same questions being asked over and over again, whereas forums help to keep q&a’s more organized so it’s easier for people to find previous answers - so I’m hoping to avoid increasing the use of slack if possible).

2 Likes

How do I join slack channel ? Where can I get an invite ?

+1 I’m also looking for the answer on how to join slack channel - and is it still alive

Waiting for a part 3 :smiley:

1 Like

No, I believe use of Slack ended with Part 1. @shriram.jaju