Viewing profile — gimagon

I can't speak to oz's error messages, but for the vega-lite + python integration, I've found altair pretty usable. In particular, simple scatter plots with tooltips were easy to ge…

Notebooks also inline visualizations more naturally than the repl, which makes them really nice for the data analysis workflow.

The definition of the Markov property is that the process is memoryless - the probability distribution of the next state only depends on the previous state. This makes them an easy…

Are you referring to the Cueva paper? I also spent a considerable amount of time trying to replicate it without success. Do you know if the author has released any code yet?

Yes, see the Hodgkin Huxley model [1]. [1] https://en.wikipedia.org/wiki/Hodgkin%E2%80%93Huxley_model

It wouldn’t really be irrefutable, it just shifts the claims of bias to what was included as training data.

His solution doesn't seem to solve the hard problem of "stuffing the ballot box." Any voter can verify their own vote is correct, but they can't verify that each vote is cast by a …

What's DSP in this context?

Could a lighter weight website serve more users for the same dollar of bandwidth as bloated website? It seems to me there's a business strategy, where rather than pushing for more …

I guess I was trying to advocate for somewhat orthogonal ideas. On the one hand, learning how to approximate functions with arbitrary precision definitely solidifies an understandi…

One small change I'd like is to teach a lot of math from a more numerical approach. Learning to approximate sines, cosines, limits, derivatives, and integrals in university made me…