Machine learning in trading: theory, models, practice and algo-trading - page 2998

[СанСаныч Фоменко]

Maxim Dmitrievsky #:
Mind blowing

Please allow out of respect.

[Maxim Dmitrievsky]

СанСаныч Фоменко #:

Please, out of respect.

Do those who write their own instruments get the "god" level?

[СанСаныч Фоменко]

Maxim Dmitrievsky #:
Do those who write their own instruments get the "god" level?

There are those who write for their own use, and there are those who write for others. It's completely different programming.

Python is for yourself. You put it out "as is". The public liked it. It is a universal language, but an interpreter, it is much more convenient than compilers at the development stage, you can do without a debugger at all.

R is for others. Thoroughly documented, input control, working out bugs and maintaining revisions. Deliberately made an interpreter, as it is intended for research. Microsoft decided otherwise and offered it for exploitation, as it turned out that R is a fast enough language, standard tools for parallelising programs with loading not only its cores but also neighbouring compotes.

[Maxim Dmitrievsky]

СанСаныч Фоменко #:

There are those who write for their own use, and there are those who write for others. This is completely different programming.

Python is for yourself. You put it out "as is". The public liked it. It is a universal language, but an interpreter, at the development stage it is much more convenient than compilers, you can do without a debugger at all.

R is for others. Thoroughly documented, input control, working out bugs and maintaining revisions. Deliberately made an interpreter, as it is intended for research. Microsoft decided otherwise and offered it for exploitation, as it turned out that R is a fast enough language, standard tools for parallelising programs with loading not only their cores, but also neighbouring computers.

Thanks, boss. Not a fan of either (r and Microsoft)

[Rorschach]

I'll put in a word for Matlab. After Matlab, python is clumsy and uncomfortable. It's like Linux after Windows, everything seems to work, but there are always some small annoying bugs. Everyone's favourite (sarcasm) spectral analysis in Python is hard to find.

But we must admit that python has become an industrial standard in machine learning, it is the future and all that.

[Maxim Dmitrievsky]

Rorschach #:

I'll put in a word for Matlab. After Matlab, python is clumsy and uncomfortable. It's like Linux after Windows, everything seems to work, but there are always some small annoying bugs. Everyone's favourite (sarcasm) spectral analysis in python is hard to find.

But we must admit that python has become an industrial standard in machine learning, it is the future and all that.

scipy.org and a bunch of other libraries on google, there are a lot of them

Mostly non-professional programmers sit here, so listening to their arguments is such a waste. To a bad dancer, as they say :)

[СанСаныч Фоменко]

Maxim Dmitrievsky #:
scipy.org and a bunch of other libraries on Google, there are a lot of them
.

Mostly non-professional programmers sit here, so listening to their arguments is such a waste. To a bad dancer, as they say :)

Here is the list. Are you kidding me? What kind of trading needs anything from this list? Although a great example for python, a bunch of something there from some skilful person.

SciPy User Manual

• Introduction
• Special features ( scipy . special )
• Integration ( scipy .integrate )
• Optimisation ( scipy.optim ize )
• Interpolation ( scipy. interpolate )
• Fourier transforms ( scipy. fft )
• Signal Processing ( scipy . signal )
• Linear algebra ( scipy .linalg )
• Problems with sparse eigenvalues in ARPACK
• Compressed sparse graph procedures ( scipy. sparse.csgraph )
• Spatial data structures and algorithms ( scipy .spatial )
• Statistics ( scipy.stats )
• Multidimensional image processing ( scipy. ndimage )
• File IO ( scipy .io )

[mytarmailS]

Guys Maybe instead of technology, it would be better to discuss ideas and the results of testing these ideas?

[Maxim Dmitrievsky]

СанСаныч Фоменко #:

Here's the list. Are you kidding me? What kind of trading needs anything from this list? Although a great example for python, a bunch of stuff from some handyman.

SciPy User Manual

• Introduction
• Special functions ( scipy . special )
• Integration ( scipy .integrate )
• Optimisation ( scipy.optim ize )
• Interpolation ( scipy. interpolate )
• Fourier transforms ( scipy. fft )
• Signal Processing ( scipy . signal )
• Linear algebra ( scipy .linalg )
• Problems with sparse eigenvalues in ARPACK
• Compressed sparse graph procedures ( scipy. sparse.csgraph )
• Spatial data structures and algorithms ( scipy .spatial )
• Statistics ( scipy.stats )
• Multidimensional image processing ( scipy. ndimage )
• File IO ( scipy .io )

Are you deliberately twisting all your posts? Is this a kind of self-aggrandisement? Please don't spoil your already bad opinion of r-users. Or tell me, who do you work for? The question was about tsos. Or rather, resentment.

[СанСаныч Фоменко]

Maxim Dmitrievsky #:
Are you deliberately twisting every post? Is this some kind of self-aggrandisement? Please don't spoil the already bad opinion of r-users. Or tell me, who do you work for? The question was about tsos. Or rather, resentment.

Yeah, well, here's a copy of that post. Where's the C.O.C.?

Rorschach#:

I'll put in a word for Matlab. After matlab python is clumsy and inconvenient. It's like Linux after Windows, everything seems to work but there are always some small annoying bugs. Everyone's favourite (sarcasm) spectral analysis in python is hard to find.

But we must admit that python has become an industrial standard in machine learning, it is the future and all that.

scipy.org and a bunch of other libraries on Google, there are a lot of them
.

Here mostly non-professional programmers sit, so listening to their arguments is such a waste. To a bad dancer, as they say :)

Good luck to you!