jointpdf

35 ms·

271.
by jointpdf 7y ago
Does anyone have insight on the extent to which interviews for ‘data scientist’ or ‘ML engineer’ roles involve Leetcode-esque or whiteboard coding exercises? What is the norm (if any) for assessing coding and subject matter competency for t
272.
by jointpdf 7y ago
So assuming this is true, what plan of action do you recommend for an “old gen” data scientist who is strong in math, stats, ML theory, R, dataviz, ETL, research, etc., but who is not by a long shot a “full fledged software engineer”? I wil
273.
by jointpdf 7y ago
> If we want to reduce our use of CO2 producing energy sources without massive declines in quality of life or vast environmental damage there’s only one choice, nuclear. Unsubstantiated claim detected. Solar, wind, hydroelectric, and ge
274.
by jointpdf 7y ago
In my experience, unsupervised learning is more broadly useful (and easier to implement) than supervised learning. Use it for data exploration, data validation, anomaly detection, topic modeling, recommender systems, cluster analysis, etc.
275.
by jointpdf 7y ago
> Probably just another fad that will prove worthless next year Sorry, my humor detector is a little out of tune this morning. This is sarcasm right?
276.
by jointpdf 7y ago
I would recommend CVAT for annotation (for images/video): https://github.com/opencv/cvat In general, annotating data for object detection or segmentation tends to be very hard to do effectively—expect low qualit
277.
by jointpdf 7y ago
Licrosoft?
278.
by jointpdf 7y ago
I wonder how kea ( https://en.m.wikipedia.org/wiki/Kea ) compared to crows in terms of intelligence. I’ve heard those things are more than capable of breaking into vehicles.
279.
by jointpdf 7y ago
Personally, I like to listen to electronic DJ mixes because a) the music flows together nicely (by design) and b) repetition paces me and gives something for my ADD brain to latch on to (like rapidly bouncing your leg) and c) I don’t get pr
280.
by jointpdf 7y ago
I haven’t seen this mentioned in this thread yet, but there are income-based repayment plans for US loans. There are different plans, but it works much the same way as above, roughly: if you earn above a income threshold (~$20k), then you p
281.
by jointpdf 7y ago
Are you sure that the causation arrow runs: high rise construction —> traffic + price increases? It seems like less of a case of “if you build it, they will come” and moreso “they are coming, so we’d better build”.
282.
by jointpdf 7y ago
Who is Uncle Ted?
283.
by jointpdf 7y ago
I think the rankings are based on the papers that people save. There is a separate page for Twitter based rankings, which seems noisier (ie top paper from the last month is about Hearthstone). It does seem that survey/review/tutor
284.
by jointpdf 7y ago
Try this for something more objective: http://www.arxiv-sanity.com/top?timefilter=year&vfilter=all
285.
by jointpdf 7y ago
Ugh, that one was auto-correct, I swear. I have no idea what’s going on at Apple’s NLP department.
286.
by jointpdf 7y ago
This is the idea behind non-negative matrix formulation (NMF). As the name implies, it forces the entries of the embedding matrices (for both the reduced document and term matrix) to be nonnegative, which results in a more interpretable “su
287.
by jointpdf 7y ago
So I’ve had this (perhaps naive and not novel?) idea for a while and am curious what people think. What if it were required by law that all surveillance video is “anonymized” in real-time via computer vision software? In other words, use an
288.
by jointpdf 7y ago
No, that was a statement of fact. The core problem is that your comment (ironically) commits the very offense that you are falsely projecting onto the authors. Namely, you are redefining concrete terms to fit the meaning that you want them
289.
by jointpdf 7y ago
This comment is just completely nonsensical and intellectually dishonest. This is not at all what this paper did, nor what any remotely credible scientific study ever does.
290.
by jointpdf 7y ago
Since one of authors posted the paper, we do know: > “ We conducted a literature search to identify papers published prior to the early-1990s that include climate model outputs containing both a time-series of projected future GMST (with
291.
by jointpdf 7y ago
Simulation is one way to produce a numerical forecast, but the prediction of any forecast model would never be the result of a single simulation run. Model predictions are always (or should always) be reported as an expected value, or the a
292.
by jointpdf 7y ago
Damn, what a toxic and misguided comment. I have a MS in Applied Mathematics and there are many areas of math (especially “pure” math) that I’ve never touched. Same goes for my professors. The field is so vast and so deep that it’s imposs
293.
by jointpdf 7y ago
Superhuman
294.
by jointpdf 7y ago
Strongly disagree with this. Chicago is one of the greatest cities (to visit or otherwise) not just in the US, but anywhere. I’ve been there over 30+ times since I was a kid and I’m going far out of my way to visit again tomorrow, it is tha
295.
by jointpdf 7y ago
Check out the CRAN task view on this topic: https://cran.r-project.org/web/views/ModelDeployment.html One dead simple way to do this (R model —> Java production) that I’ve done in the past is to use PMML (via
296.
by jointpdf 7y ago
This sounds pretty awesome. Do you have any write-ups that explain more?
297.
by jointpdf 7y ago
Another great one is this documentary of the Fitz Roy Traverse by Honnold and Caldwell. It’s a high alpine adventure rather than a free solo, but it’s a different sort of amazing achievement to watch unfold. https://www.redbull.c
298.
by jointpdf 7y ago
There’s actually an entire book on this at: https://10print.org/
299.
by jointpdf 7y ago
Here’s a pretty comprehensive answer on the topic from the original UMAP author: https://github.com/lmcinnes/umap/issues/25 Clustering the output of UMAP is also given a nice tutorial in the docs: https:&#x2
300.
by jointpdf 7y ago
If it runs Python, then couldn’t you use SymPy as a CAS (in addition to NumPy, seaborn, etc. for numerical computing / visualization)? By the way, I like the idea of this calculator and hope to see more like it. I think something that

More ›