stealthcat

13 ms·

91.
by stealthcat 9y ago
How horrible is horrible
92.
Ask HN: GSoC experience?
3 points by stealthcat 9y ago|1 comments
93.
by stealthcat 9y ago
What does it mean by modern?
94.
by stealthcat 9y ago
seeing that Pytorch started as Chainer 'fork' (?) with similar API but different backend, showed us how much a BigCo can do against a startup, it's unsettling. Chainer autograd has been around years ago.
95.
by stealthcat 9y ago
> It must be beyond our scientific knowledge https://www.youtube.com/watch?v=PVCbsZtB4-4 The problem of accepting revealed knowledge. Whether to hunt down the best one or settle with assumptions. The common trap for many
96.
by stealthcat 9y ago
Buy him a desktop with a big screen. Put it near kitchen or dining table.
97.
by stealthcat 9y ago
Seems like me. I am studying in Tokyo, Japan. The pressure is so high as I am surrounded by highly-intelligent and hard-working people around me. The machine learning research for Master's degree here is somewhat equivalent to PhD leve
98.
by stealthcat 9y ago
What happened to bioinformatics?
99.
by stealthcat 9y ago
Because Linux font rendering is awesome. MacOS is about as good. Windows Cleartype is not even close. You are going to read code all day -- that comfy feeling when reading them is important
100.
Advice on Reading a Book(youtube.com)
1 points by stealthcat 9y ago|0 comments
101.
by stealthcat 9y ago
An extreme example would be porn industry. No benefits but toys are really profitable to be simply let go.
102.
by stealthcat 9y ago
A Life Worth Living https://www.youtube.com/watch?v=ssc_dJjcnv8
103.
by stealthcat 9y ago
Those are undergrad courses. PhD should already know those by heart.
104.
by stealthcat 9y ago
>If you are doing a laundry cycle and you have to wait for the dryer to to finish before you load the washer its going to take lot longer to do the laundry. Reminds me of that lecture notes slides from MIT OpenCourseWare on digital elect
105.
by stealthcat 9y ago
If I were to start such service today how should I do it? Is SEO the only way? Now if I google "temporary email", top results rarely changes
106.
by stealthcat 9y ago
How is your startup doing? Can you share your story? Maybe post in IndieHackers? I am interested in freelancing/consulting in data science and deep learning but don't know where to start.
107.
by stealthcat 9y ago
Well it's your data, you could do some data science and NLP to study your life since 20 years ago.
108.
by stealthcat 9y ago
Thanks for reply. Since I am learning deep learning, I still need to read and run Lua Torch code every now and then. But if I have choice I choose Pytorch. P.S. I did not downvote you. Somebody else did
109.
by stealthcat 9y ago
Prototyping is what you usually do in ML/DS. You code incrementally and break things really fast. Python (IPython/Jupyter) and R's interactive mode is really indispensable for this. Without restarting/re-initialize every
110.
by stealthcat 9y ago
Convince me to Lua. While Lua gets job done, as an outsider who dabbled into Lua (Torch, Love2D, ESP8266) I think Lua ecosystem have two major problems: - 5.1 vs 5.2 vs LuaJIT. This is worse than Python 2 vs 3 - Lack of universal standard o
111.
by stealthcat 9y ago
https://www.reddit.com/r/MachineLearning/comments/7lb5n1/d_c...
112.
by stealthcat 9y ago
Chainer does have its tradeoff though. It's not exactly smooth to generate CUDA kernels from pure Python code. Also Chainer is bogged down by the fact it is made and maintained by Japanese company/people or at least majority of th
113.
by stealthcat 9y ago
They teach you how to efficiently use (two kinds of) hammers and what nails to knock 1. ConvNet (for images and structured data) 2. RNN (for text and sequence data). They don't teach you far advanced stuff, like how to creatively misus
114.
by stealthcat 9y ago
1. Get yourself grounded in theory Learning from Data by Yaser Abu-Mostafa, and its companion book. Very theoretically heavy but worth the trouble (this is Caltech course) Do all assignments in the course with Python/Numpy/Scikit-
115.
by stealthcat 9y ago
Always stir fry some finely chopped onions and garlic before putting anything else
116.
by stealthcat 9y ago
TF has TF Eager in latest version (1.5) Though when comparing a same model code written in Pytorch, TF Eager imperative code feels verbose and bit cluttered as if it is written in Java.
117.
by stealthcat 9y ago
How to use this book? My machine learning lab have few copies but nobody seems to be using it.
118.
by stealthcat 9y ago
Suggestion: Put registration forms in the main page. Put simple captcha. Remove whatever barriers to encourage registration
119.
by stealthcat 9y ago
What about JSX? Hyperscript's bracket closing is quite tedious to me.
120.
by stealthcat 9y ago
Nim

More ›