closed

18 ms·

151.
by closed 7y ago
> ... ask about arrests and not convictions Expungements are for clearing arrests[1] > or an employer fires you for a conviction found on a third party database what are you going to do? Ask for a copy of the report (they are required
152.
by closed 7y ago
They failed to replicate the study over multiple attempts, 1 with four times the sample size. Using the time reversal heuristic, if we flip the order, so the replication came before the original, how would you feel about Science publishing
153.
by closed 7y ago
Is this true? Couldn't we run a process on top of a lot of humans, which, assumming they are weak learners, garuantees strong learning (i.e. boosting)?
154.
by closed 7y ago
These days, I often use Rmarkdown for both python and R work. On the R side, Rstudio makes it a pure delight. On the python side, jupytext makes it easy to open / edit Rmds as notebooks. Hats off to the many people in these communities
155.
by closed 8y ago
I don't think that's the correct use of prop.test for this question. When you give it two numbers (success, total), it tests against the null that chance of success is 50%. Here, we want to test whether p(success|cond) differs acr
156.
by closed 8y ago
I would say hands down, plotnine. It... * is a port of ggplot2 to python * has a very active maintainer I've been testing out porting a bunch of analyses from R to python, and it has been very easy to swap in for ggplot2. The biggest d
157.
by closed 8y ago
You can use a correlation matrix instead of covariance. If I understand correctly, the problem you're describing also occurs is you try to interpret a linear model without normalizing first (for the same reasons).
158.
by closed 8y ago
It sounds like you're saying that acceptance criteria are hard to define when there's high uncertainty (a lot of activities are high risk of not yielding anything interesting), which makes sense. In that case, it seems like... * a
159.
by closed 8y ago
This is really helpful article, and I think the concrete questions they lay out are ones all PhDs should answer explicitly when starting a new project. That said, I think the biggest risk to PhD research is not just project, but product man
160.
by closed 8y ago
Ah thanks for pointing out--I mostly agree with his posts (and his minimally sufficient pandas is a great one!), and it's definitely worth reading. A common quirk with a lot of the python DS books is them being "reference manuals&
161.
by closed 8y ago
I have a fair amount of experience with pandas, and find the notebooks very help to refer to! I would say it's worth noting that his book is organized by technology (e.g. numpy, then pandas, then plotting), which makes it feel more lik
162.
by closed 8y ago
There's also jupytext, which provides round trip conversion between Rmarkdown and jupyter notebooks, so you can edit the markdown in jupyter, or in an editor and have it automatically converted to ipynb!
163.
by closed 8y ago
Any area of statistics that does sequential sampling can be framed as RL. Two areas that stand out to me are all non trivial forms of A/B testing and adaptive (educational) assessment.
164.
by closed 8y ago
In Rmarkdown you can set cache=TRUE for the code chunk you don't want to rerun, and it won't rerun it unless something above/in it changes.
165.
by closed 8y ago
I see method of loci more as "ram", whereas spaced repetition is moving things into a hard drive. For example, when reading, I'll often use a mental room with 10 spots x 3 key points per spot to keep an outline. Periodically,
166.
by closed 8y ago
Have you tried a strategy like method of loci? If not, I would try it first, to see how it goes. Visualizing is one part of the strategy (depth of processing), but a huge other piece is that you are enlisting spatial cues to help prompt rec
167.
by closed 8y ago
This is a classic discussion in the intelligence literature. For example, the cattell-horn-carroll theory[1]. Usually meauring general intelligence involves measures along a bunch of different problem domains. [1]: https://en.m.w
168.
by closed 8y ago
Thanks for your response--it's helpful to hear! I've mostly been using flask for simple APIs / dashboards, but have wondered lately whether a more "batteries included" approach like Django could be helpful once a db
169.
by closed 8y ago
When have you been using Django compared to Flask? I'm curious, since I often use Flask + vue, but haven't used Django much.
170.
by closed 8y ago
It's worth noting the second can't be used inline. I don't know the history, but it seems like that might be one of the reasons `x() if y()` isn't allowed in python. [x() if y()] if it existed, might carry an explicit el
171.
by closed 8y ago
Useful review. One caution I have while reading posts like this, is that the author does not distinguish between the testing effect, and the spacing effect. While the value of testing (and especially spaced testing) is well established [1],
172.
by closed 8y ago
> I regularly wish python required some sort type indication for function parameters You can use type annotations, either via the builtin lib or 3rd party libs! https://docs.python.org/3/library/typing.html &gt
173.
by closed 8y ago
Peeking at your data, and calculating the sample size you need for a test are separate statistical issues. I agree that peeking messes up significance levels :). The point I was trying to make was you can decide to run a test with a very sm
174.
by closed 8y ago
I don't understand why you would need a sample of a certain size. Setting a significance threshold at 5% takes sample size in to account. For example, if I ran a permutation test with a sample size of 5 in each group it could never bee
175.
by closed 8y ago
The hardest problem I've ever solved is learning to sit down and collage. Like, where you cut out bits of magazine and paste them onto paper, etc.. My friends do it pretty often. I find it super relaxing, and know I'll be happy I
176.
by closed 8y ago
Wow, your weren't kidding! I did a deep dive to compare airtable to a lot of other planning tools and never even noticed Notion... The one downside compared to airtable seems to be no automatic backups. But the ability to format notes
177.
by closed 8y ago
I suspect that much of this issue is resolved by looking at how they sought consent for their study, since they had to inform participants about what they were going to do to them (though they could have phrased it as, I may do A, or I may
178.
by closed 8y ago
RE python in Rmarkdown being limited... > The Python session ends after the cell executes, making it unhelpful for tasks other than ad hoc scripts. This isn't the case anymore thanks to the reticulate library. https://blo
179.
by closed 8y ago
Of note--while Rmarkdown's python support used to be pretty bad, the most recent Rstudio uses the package reticulate behind the scenes, and it works really well! It's easy to share data between the R and python session, and callin
180.
by closed 8y ago
Can and does! Jupyter notebooks have a flexible architecture for running anything (called "kernels"). Notable examples... R: https://github.com/IRkernel/IRkernel node: https://github.com/notab

More ›