Why is it so sensitive to word order?
Esp8266 USB input Manufacturing: Bad: 0.0775 - Good: 0.9290
Manufacturing Esp8266 USB input: Bad: 0.9992 - Good: 0.0009
Show HN: Testing HN titles against a neural network
221–225 of 225 posts
Re: Show HN: Testing HN titles against a neural network
#222I'd like to see HN add a classic tool: the kicker (https://www.easymedia.in/kickers-newspapers-use-even-today/ ). "It provides [headline writers] the extra space that they desperately need to pack meaning in headlines."
This could easily be implemented as a (hover) tooltip.
Re: Show HN: Testing HN titles against a neural network
#223Earlier quoted context omitted.
What tools would you use to detect overfitting in this case and in general?
My brain. You will overfit an NN trained on only 1000 examples. Also a simple train/test split will tell you that. But the author failed to take any time to learn the basics before spewing out this drivel.
By now it's clear from your history of behaving aggressively in HN threads that you don't mean to use the site as intended. I've banned the account. https://news.ycombinator.com/item?id=21245546
I'm sad to do that because you are knowledgeable about a number of things. Many of us could learn from you if you would share what you know without putting other people down. But the aggression subtracts more than the knowledge adds. We can't have users behaving like this sort of asshole in comments, least of all in Show HN threads, where the idea is to teach people things and expressly not to shit on them and their work: https://news.ycombinator.com/showhn.html.
Other users here know things and are willing to talk about them without being mean. GistNoesis modeled this wonderfully in the GP comment. We'll learn what we can from them instead. But if you decide that you want to use HN in the intended spirit, as described in the site guidelines and especially the one that says Be kind, you're welcome to email hn@ycombinator.com and let us know.
Re: Show HN: Testing HN titles against a neural network
#224Congratulations on getting first rank on front page. Congratulations on getting your hands dirty and doing everything yourself like computing gradients manually, badly shuffling (non Fisher-Yates), badly js transpose (double swapping), it is a great way to learn. Congratulations on completing a full pipeline, that's the hard part then it just swapping pieces for better pieces. I advise non-technical readers not to at…
I really like this comment. Thanks a lot for taking the time to going through the code and figure all that out. The way I learn things is by doing. Since I started college, the only way I can understand a mathematical equation is by turning it into code. The machine learning class I took was a basic one. We learned about knn, kmeans, adaline, perception, linear regression... the semester ended with a multilayer perceptron with only one hidden layer. This Dejavu NN you see in my code is like the fourth iteration of me trying to really understand NNs. Some of the things you said I already knew but didn't took the time to apply: like cross validation. Tht shuffle algorithm I didn't knew, thanks for that. Anyways, if all the criticism I got from my projects were like yours i'd be such much better now.
When I learned about KNN, I made this project: https://github.com/victorqribeiro/budget
When I learned about Kmeans, I made this one: https://github.com/victorqribeiro/groupImg
When I learned about Perceptrons: https://github.com/victorqribeiro/carGamePerceptron (the training is done when you run the project)
When I learned about MLP: https://github.com/victorqribeiro/jokenpo
When I learned about Neural Evolution: https://github.com/victorqribeiro/aimAndShoot
None of these projects are perfect, but helped me to better understand things. I did implemented a professional solution for a company using machine learning, but I used scikit-learn (back then). I have some experience with tensorflow also, but implementing things makes me fell like in control. But I know whena a solution is good enough for production and when it's not. I made clear on the README of this project that this is not at all good for production. I took the time to show every single step of the process and tell it how it is. I had limited time and limited resources, and with the little I had I made a "fun" experiment. I enjoy doing this kind of thing. I have to squish this kind of projects in my free time, cause I have a full time job and I'm finishing my final thesis (I'll defend it two week from now).
Sorry if I turn this into a journal entry, but I kind got hurt yesterday when the other guy shted all over my project That's not at all what you did and I appreciate that. Thanks again.
Re: Show HN: Testing HN titles against a neural network
#225Well, here's the thing: a good Samaritan offered 2.6M stories from HN with score. I've downloaded the file (almost 500M) and I'm now processing it. It is taking a long time to just process it. I don't know if I'll be able to train the neural network with all that data. As I said on the repo, the project is was a quick thing, just to test a theory. My question is: do you think is worth feed the NN more data so it can…
Depends - if you really want to explore the tech, this is precisely the way to do it. I would be interested in the results, especially comparing them to your initial results.