These are chat archives for FreeCodeCamp/DataScience

28th
Oct 2017
Alice Jiang
@becausealice2
Oct 28 2017 05:18
@evaristoc I hadn't heard of it before, but glancing over their website it seems worth exploring. Why do you ask?
Alice Jiang
@becausealice2
Oct 28 2017 05:42
Is anyone here better than I am at D3? The D3 gitter is dead, the Slack regularly forgets about me, and I've run into a logic roadblock :(
Josh Goldberg
@GoldbergData
Oct 28 2017 06:28
Hello…new to the chat. Hoping it’s as fun and responsive as the other FFC chats. Anyone use R here?
evaristoc
@evaristoc
Oct 28 2017 09:21
@becausealice2 just let the question here? I would be happy to give it a go. Need a refresh with D3 and helping you could be the sort of challenge I need to revive my D3 skills. Hope it is ok?
@GoldbergData We are a few here in this chatroom. It might take some time to come with an answer. R? Several: @erictleung and @mcbarlowe are currently the most active. I know also R but I am more into Python.
evaristoc
@evaristoc
Oct 28 2017 10:12
@GoldbergData what do you exactly want to know regarding R?
evaristoc
@evaristoc
Oct 28 2017 10:45

@becausealice2 You asked about my interest in datadotworld. I am currently exploring a proposal to make use of this repository: https://github.com/freeCodeCamp/open-data .

One of the things I am evaluating is how to store datasets that are too big to be kept in the Github repository. datadotworld could be an option.

The repo is planned to host past, present and future projects that have used fCC data. That includes your projects, @becausealice2.

Alice Jiang
@becausealice2
Oct 28 2017 11:26
You mean my map that's very, very broken that never got finished?
Speaking of maps, the D3 problem I'm having is that I have a map with franchise locations that I am trying to add a slider to so when the user increases the year with the slider, it shows which locations open that year, as well as keep the ones of previous years on map
I don't know if I'm making sense and my GitHub repo is very behind. I'm not at my laptop or I'd push commits and show you :/
evaristoc
@evaristoc
Oct 28 2017 11:33
@becausealice2 that and the scraper.

People

Although I can solve it myself, I would like to know from some of you if you have worked on Restricted Bolztmann Machine and where to use them?
Seems to be a simple design...
Also, who of you have been working with Autoencoders?
evaristoc
@evaristoc
Oct 28 2017 11:46

For those who are interest in Deep Learning and Machine Learning in general:

The big topic seems to be UNSUPERVISED LEARNING. Not that you have to focus on that, but keep that in mind.

The big, important contribution that developments in unsupervised models will bring is to allow the system to learn from scratch without any previous data or no much data.

Then think about the following: how to formulate a new emerging learning by combining different learnings. Think of the semantic web. This is something that current system are not doing properly.

This reminds me an article I was recently reading about AlphaGo. A recent development of the DeepMind's AlphaGo program put in practice an unsupervised learning approach : instead of using previous learnings (games) played by humans of the the Go game, they put the new version to play against itself, learning to play Go from scratch following a set of rules. The final results were:

  • it took less time for the algo to learn
  • it developed a larger range of strategies to become unbeatable

https://deepmind.com/blog/alphago-zero-learning-scratch/

evaristoc
@evaristoc
Oct 28 2017 11:57

What is the conclusion? MODELS are still very important. In this case, the model you are providing (the set of rules of the game) allows a good level of development of the program.

Machine Learning is based on DATA but it is not able to provide more answers beyond the data that the algorithm gets. Machine Learning is just an optimisation heuristic more applicable when you don't have that model by hand, but if you compare that with for example the impact of the Einstein's theoretical work, I think you would agree with me that the impact that Einstein had on providing explanations about the surrounding world is still more valuable because you can derive new knowledge from it.

That it is why Unsupervised Learning is so important: they are trying to create a computer that it is able to think of new solutions, not to optimise the existing ones.

If you manage to find the MODEL that explain the phenomenon (think of maths here), you win more. So my advise is: keep working with models. Machine Learning won't exclude the needs of Math, I am afraid.

@samosale!!! We are just talking about the new repo I am trying to work on... I expect to include your work there too... The 3D map.
Nice to see you around.
Aleksandar B.
@aleksandar-b
Oct 28 2017 12:04
@evaristoc nice to see you too :+1: That's great, i will have some free time in 2-3 weeks from now.
Aleksandar B.
@aleksandar-b
Oct 28 2017 12:16
I must say that I understand very little of what you guys are posting here. Data science is so big and complex field right now. It will took me probably 1-2 years just to learn all this. Impresive field.
Josh Goldberg
@GoldbergData
Oct 28 2017 17:23
@evaristoc I don’t want to know anything in particular. I was just curious the R population in the chatroom
Alice Jiang
@becausealice2
Oct 28 2017 21:02
That scraper :disappointed:
What a mess. I hate Node so much :joy:
@GoldbergData I use R when I have to
Josh Goldberg
@GoldbergData
Oct 28 2017 21:07
I was learning python. But got a job where they use R. So I had to switch.
I’ll revisit python soon enough.
evaristoc
@evaristoc
Oct 28 2017 22:10
@GoldbergData what kind of job you are doing?