abakus

12 ms·

61.
by abakus 8y ago
The proliferation of yaml is rather unfortunate.
62.
by abakus 8y ago
They usually cannot improve carefully optimized numpy operations, nor GPU operations. The use cases are thus limited.
63.
by abakus 8y ago
This. A lot of GPU deep learning libs depend on CUDA, cuDNN, which is not solved by pipenv / virtualenv, BUT is actually handled by conda.
64.
by abakus 8y ago
Merit, tech skill based please.
65.
by abakus 8y ago
But the whole point of this discussion is that you should probably duplicate as many times as needed if it helps to avoid wrong abstraction.
66.
by abakus 8y ago
It is almost a non problem for Chinese text: words are mono spaced squares.
67.
by abakus 8y ago
I know the Dark Matter Theory is very successful at explaining lots of phenomenons of cosmological physics. Nevertheless, it always reminds me of the Aether Theory.
68.
by abakus 8y ago
In contrast, the supply of shaolin monks has always been steady
69.
by abakus 8y ago
Also they tend to downvote comments that do carry weight, so from this perspecitve, they are not that positive after all.
70.
by abakus 8y ago
THis. I think the type hint system adds so much visual noise it starts to look like perl. When an attempt to build huge (100k+) project with Python failes, it is not because of the duck typing, but simply because Python is not designed for
71.
by abakus 8y ago
Apparently everybody knows "correctness" comes first. But that is the whole point of this article: It argues that complexity will reduce correctness, and thus simple should come first. Please NOTE that I am not saying that I agree
72.
by abakus 8y ago
Code with a mix of Y% #$ and lower_UPPER_CamelCase looks ugly.
73.
by abakus 8y ago
Most scripts do not have reusability and not warrant sharing. If they do, make them into a library.
74.
by abakus 8y ago
Talk is cheap, show me the prototype?
75.
by abakus 8y ago
but y is uninitialized, am I missing something?
76.
by abakus 8y ago
WTF is this "for (x=y; ...)" part?
77.
by abakus 8y ago
It doesn't matter whether we are alone or not. The speed limit is so slow, the dilation is so quick, we will never be able to contact.
78.
by abakus 8y ago
Why United States doesn't dominate soccer?
79.
by abakus 8y ago
I use Numpy NDArray and pandas DataFrames a lot, and often times you want to validate the shape / columns / dtype of theses objects. I have yet to find a clean and meaningful way to type hint these, and ended up just put assertion
80.
by abakus 8y ago
One thing I always dislike about julia is the .+ .* .< ... all those dots for element-wise operations. It is inherited from Matlab, but is a bad design because more common usage should have shorter operators.
81.
by abakus 8y ago
I like the spirit of it, but the implementation seems to contain too many exceptions (e.g. trailing comma, whilespaces around :). This problem is not unique to Black, it is actually too common for most auto-formatters. I usually prefer a se
82.
by abakus 8y ago
I second that. They are avoiding the real problem by pretending to fight for social justices.
83.
by abakus 8y ago
The premise that we think we can pack an infinite amount of information into finite space time just feel wrong to me.
84.
by abakus 9y ago
But that is still 2 keystrokes. That is why in R they came to the compromise to use both <- and =.
85.
by abakus 9y ago
The length of symbols should roughly be negatively correlated with its frequency. Since in procedural language, assignment is a much more common operation than equality check, it is reasonable to favor "=" over ":=" or e
86.
by abakus 9y ago
I have no problem with the methodology and results of this article, but I think conclusion that "functional languages are better than procedural languages; ..." only applies to bugs; in reality the choices of programming languages
87.
by abakus 9y ago
The biggest problem with R: it is too slow.
88.
by abakus 9y ago
Text editing efficiency is overrated, at least for coders. thinking >>> editing. If serious non-trivial editing effort is required, write a script to do it.
89.
by abakus 9y ago
SGD > adaptive, according to this: https://people.eecs.berkeley.edu/~brecht/papers/17.WilEtAl.A...
90.
by abakus 9y ago
Wait，Java has actually the shortest silhouette?

More ›