Is it possible to open source the source with the model and the training data?
Show HN: Testing HN titles against a neural network
191–200 of 225 posts
Re: Show HN: Testing HN titles against a neural network
#192Very cool. I had the same idea a couple of years back and implemented a very similar interactive tool [1]. If you find the topic interesting, then you might also enjoy the analysis explained in that blog post.
Re: Show HN: Testing HN titles against a neural network
#193Earlier quoted context omitted.
So the answer "Can a neural network predict if your HN post title will get up votes?" is a clear "no", at least for this tool.
The only one it predicts well...is itself.
Perhaps because the author had access to the tool before writing the title? In which case, less of a "prediction."
Re: Show HN: Testing HN titles against a neural network
#194How many Hacker News stories was this network trained on? From the code, it wasn't many, and you need a lot of stories.
A year ago I made a Hacker News submission score prediction notebook on the full HN corpus: https://www.kaggle.com/minimaxir/hacker-news-submission-scor...
Even with hundreds of thousands of data points, the R^2 was effectively 0.
Re: Show HN: Testing HN titles against a neural network
#195Congratulations 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…
This. Exactly this. No sophisticated tokenization. No interesting architecture using attention. And the author is completely clueless about overfitting... and even cross entropy loss. He could have gotten better results just using a bag of words approach.
But this ends up on frontpage anyway. Welcome to HN.
Re: Show HN: Testing HN titles against a neural network
#196This enforces the well-known rule that, if an article asks "Can...?" or "Does...?", the answer is always 'No'.
Re: Show HN: Testing HN titles against a neural network
#197Current top 10: 1. "Apple introduces 16-inch MacBook Pro, the world’s best pro notebook" Bad: 0.9964 - Good: 0.0038 2. "Developing open-source FPGA tools" Bad: 0.3381 - Good: 0.6652 3. "Show HN: Can a neural network predict if your HN post title will get up votes?" Bad: 0.0598 - Good: 0.9307 4. "How internet ads work" Bad: 1.0000 - Good: 0.0000 5. "More Intel speculative execution vulnerabilities" Bad: 0.7413 - Good:…
"Show HN: Cow Robots Neural Network Smells"
Bad: 0.0267 - Good: 0.9720
Re: Show HN: Testing HN titles against a neural network
#198Anything that doesn’t factor in the time of the posting will be pretty much useless
Re: Show HN: Testing HN titles against a neural network
#199Well, 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 make better predictions? Please up vote this comment so more people could give their opinion.
Re: Show HN: Testing HN titles against a neural network
#200"Google Uber for Carbon Sequestration" Bad: 0.1631 - Good: 0.8615