Every incline is followed by a cliff
Very frustrated, since the dataset still seems very noisy. DAMMIT! https://bit.ly/3eKdnzW
Posts by Ramil (142)
Amazing progress!
I have decided to pull back a little on my notion of doing image clickbait classifier, and instead decided to focus on text for 3 reasons: 1) It's more straightforward 2) Datasets are more available 3) There is a lot of academic work done And after spending the whole day on it, I have developed a VERY good model. The trick, of course, was to write as little code as possible since it then forces you to write not only more optimal code but also to not reinvent the wheel. I'm actually super impressed by how the model is performing! Over 99.5% accuracy on the trained dataset with training taking just a couple of seconds. WOW! Results uploaded to github!
Interesting
Had a short chat with Lex Fridman. Didn't realize that he was even more down-to-earth than he already seemed to be. Geometric Side Note: I would say Lex is a midpoint between me and Elon
Clickbait Project Update
Did a little more (boring) manual data labeling, and then developed a ML model that slightly outperforms naive algorithms. While results aren't that impressive, I'm glad to have gone from 0 to ... epsilon (math joke intended) all on my own applying the recently acquired data science skills
Weekly Update!
Couple of things: 1. Performed much better than expected in Google's Kickstart getting 109th place in the US! 2. I'm getting a better idea of what I would include on Competitive page. Probably going to put my contest ratings there, like the plot in the sidebar and have a list of templates for coding (like optimized int read, setting higher recursion limit, etc.) 3. Since I'm done with TF course, I should probably find another one. After quickly looking through the options, didn't find anything appealing, so gotta spend more time on it. 4. Also should restructure some database stuff on the website, maybe add an admin panel, and change this mediocre font to a more appealing one.