Machine learning in trading: theory, models, practice and algo-trading - page 355
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Source code of the package. Download, fix, compile. Sometimes this works, sometimes not. Maybe there are only 2 lines to correct, or maybe a lot).
SZY The easiest option, download the previous version of R.
install.packages("drat", repos="https://cran.rstudio.com")
drat:::addRepo("dmlc")
install.packages("mxnet")
it installs like this )
Remains to figure out what to shove input and output and you can have fun.
If I could just figure out how to work with R in visual studio, that would be even more interesting
Calibration makes a "hard" classifier into a "soft" one ......
And if you give it porn files to analyze, it becomes hard again...
If we could figure out how to work with P in Visual Studio, it would be even more interesting
This is a question about R, how do I make the versions compatible?
It used to be possible. Both on R and on the Internet there are plenty of copies. A couple of h-files and you're up and running. If you do it the old-fashioned way, you should put the Rcpp package.
If you do, please share your experiences. This is interesting. I can't figure out how RStudio is better.
If you install it, please share your experiences. I wonder. I cannot understand how RStudio is better.
Rstudio, too, did not understand what it gives. Took off.
With MS-R - I don't know what kind of thing. Are the charts scalable there? The most uncomfortable thing in R - the impossibility to stretch the graph without rescaling.
PS I looked at MS pictures of R in VS. Very decent. Wanted to put it, but VS 2017 crashed, apparently with another update. Tried to fix it, not fixed - kicks out.
Uninstalled. Paragraph VS2017 is no longer installed. Installer gives an error. Here we go.)
An article on LSTM is in the plans.
I'm going to finish up an article on the new version of darch_0.12 (I promised) and after that two articles:
1. Machine Learning Automation (using TPOT::Python).
2. LSTM - how it works (R/Python)
I'm desperately short of time
install.packages("drat", repos="https://cran.rstudio.com")
drat::addRepo("dmlc")
install.packages("mxnet")
it installs like this )
It remains to figure out what to put on the input and output and you can have fun
I wish I could figure out how to work with P in Visual Studio, it would be even more interesting
The main difference of mxnet from other R machine-learning packages is that it was developed for Python and later it was adapted to R API. Since R data is "column-oriented" (colmagor) and Python data is "row-oriented" (rowmagor)
data preparation is specific. In latest version mxnet_0.9.4 this seems to be automatically recognized, but I have not really checked.
By the way, there are a lot of examples of applications.
What's wrong with Rstudio? It is a fully functional IDE which is developing and well supported.
If you feel comfortable there, though.
Good luck
Rstudio, too, did not understand what it gives. Took off.
With MS-R - I do not know what kind of a thing. Are the charts scalable there? The most uncomfortable thing with R is that it is impossible to stretch the graph without rescaling.
1. Rstudio allows you to do the whole cycle of work from writing and debugging scripts to preparing documents.
What's wrong with graphs? What kind of scaling are we talking about? Let us know, maybe we can give you a hint.
Good luck
You have to put the R that is from Microsoft - MRO Microsoft R Open. They are behind the times, but they also track package versions. Then you can lock yourself in and fence off progress.
The MPO option is good.
To fix packages and their versions used in scripts, I use their same package checkpoint/ / Works fine.
Good luck