65

After months of tedious research, I finally feel like I understand machine learning.

All of my programmer buddies are in envy, but I keep trying to explain that what I finally get is that it's not as hard as it's presented to be.

I feel like a lot of the terminology in machine learning is really pretentious and unnecessary, and just keeps new people from the field.

For example: I could say: "Yeah, I'm training a classification model with two input neurons, a hidden activation layer, and an output neuron", and you might think I was hot shit. But that just gets translated into "I'm putting in two inputs, sorting them, and outputting one thing".

I feel like if there was a plain language guide to machine learning, the field would be a lot more attractive to a lot more people. I know that's why it was hard for me to get in. Maybe I'll write one.

Comments
  • 9
    Might be a good thing to make then. The more people that will use/understand it the faster/better it wil grow. From wich we all benefit.

    Its still on my if I ever find the time to do it list. I work a lot with analist that search for paterns and abnormalities in data. And I feel that their is a lot to gain there.
  • 5
    @willbedow if you want to write a couple blog posts about this with me, I'd be very happy. But do you understand backpropagation? I feel like I sort of know how it works on a high level, but I've never taken time to look over the equations and get a strong intuition.
  • 2
    @thejohnhoffer My understanding of backpropagation is pretty high level too. But I'm definitely up for writing something about it.
  • 2
    @willbeddow awesome! Let's do it. What's your Gitter? I'll add you to gitter.m/learn-blog
  • 4
    Where did you started your research with machine learning? I am about to start getting in to it as well
  • 10
    I like people like you. Humble. In fact most if not all of this community is humble. I love devRant.
  • 1
    @willbeddow I meant https://gitter.im/learn-blog for the group I just created
  • 2
    Good idea! I'd definetely read it
  • 1
    @kjone I actually don't remember where I started, but I remembered that it was really frustrating. Where I'd recommend starting is with keras.io. It really simplifies the concepts and puts it in terms of normal programming for me.
  • 2
    @thejohnhoffer Just made a gitter account. I signed up with my github, so it should be ironman5366
  • 0
    @willbeddow nice going out for you. But how did you implement any of this stuff on python I have been learning it only in octave using the mathematical equations and that too just for starters like linear regression, logistical classification, regression analysis.
  • 0
    @sam9669 keras.io has been my savior. I don't have to know any of the math, I can just write something like model.add(Activation('softmax')). It's amazing. Which is great for me, since I'm definitely not proficient enough in linear algebra to implement it myself!
  • 3
    I'm setting up a github pages blog for it since you guys seem to be interested, it'll be at https://ironman5366.github.io/learn... once I get it running. If any of you guys want to help, I'd love to collaborate with you. Shoot me an email at will@willbeddow.com and I'll add you on the project.
  • 1
    @willbeddow Awesome! I'll send you my email!
  • 1
    @thejohnhoffer @aronmik please make that blog 😂
  • 4
    @n1had Just made it! Working with @thejohnhoffer on it as we speak. You can view it at https://ironman5366.github.io/learn... or join the gitter at gitter.im/learn-blog
  • 2
    @willbeddow awesome! But that font dude lol
  • 0
    Well.. seems ur work focussed on neural networks as u said u started wid keras. I would say many people are already into ML and half of the crowd r in a state where they relate every other problem statement to ML.
  • 2
    @kjone sharing my experience.. I would say start from Coursera Machine Learning course by Andrew Ng. It's a great start towards ML n lay down a very strong foundation. It covers variety of topics and explained both mathematical and programming.
  • 2
    @timekeeper I actually didn't start with keras - I wish I had XD. Keras is kind of the magic module that made it all make sense to me, but I didn't find it until a few months ago. I started with resources like tensorflow, theano, torch, etc. and by reading whitepapers on the topic.
  • 0
    @willbeddow cool.. u have played wid a lot of libs.. did u got ur hands dirty wid coding tf to run in a GPU?
  • 1
    @timekeeper It actually wasn't that hard. I just had to install tensorflow-gpu and the CUDA libraries and then everything happens automatically. And it even works when I'm using keras on top of tensorflow
  • 1
    @willbeddow awesome.. I am waiting to get a Nvidia soon.. Currently surviving with scikit and xgboost...
  • 2
    <bookmark/>
  • 2
    Commenting so I can read the blog later;
  • 0
    Anyone who wants to help with the blog is welcome to join http://gitter.im/learn-blog -- :)
  • 1
    I conspire that all this jargon is to keep competitors away. Pussy tactics
  • 1
    It took about 40 YouTube videos and 3 hours reviewing code that people uploaded on github to get what Ai is. I'm 16 by the way and although i haven't even finished highschool math, I don't think its that difficult of a concept to grasp.
Add Comment