I need to make a neural network to fit some data, a very large set of data, and my laptop is not powerful enough to use the training tools in the Neural Network Toolbox. I have access to a nice Linux cluster with Matlab, but the compute nodes don't do X11, so I can only use command line or script.
After reading the Mathworks page on Using Neural Network Toolbox, it looks like it should be possible to harness the functionality of tools such as nftool from the command line, but even after looking at the Getting Started Guide and the User's Guide, I found no understandable instructions of how to start calling commands to the NNT without bringing up a GUI.
The closest was on page 20 of the User's Guide, in the third paragraph, where it says that reading the code of the GUI tools provides a good introduction to using the toolbox by command line. However, the code (found in C:\Program Files\MATLAB\R2016a\toolbox\nnet) contains a lot of references to commands I'm not familiar with, particularly running Java from within Matlab.
Despite having used the nftool several times, I do not recognize any of the lines as "Ah, this is what does that". So, I guess what I'm looking for is a set of instructions that connect the commands to what they do.
If you use the NN UI first, it can generate the matlab code for you, as an example. There is no simple answer to your question as the NN toolbox has quite a large array of functionality, so essentially the answer would be a complete tutorial.
http://www.mathworks.com/help/nnet/gs/fit-data-with-a-neural-network.html#f9-33554.
In fact if you click that link, the advice is precisely what I was saying above.
After stepping through the nftool (or similar) screens, don't stop at the Deploy screen, keep going by clicking next to the final 'Save' screen and choose simple or advanced scripts.
Related
I'm wanting to make an application with node have a cli interface, as it needs to be ran in a terminal. I want to split the terminal into several sections, one with some identification as to who is viewing the application, another with some other random info, a menu on the side that you can use the arrow keys to move up and down the options, a main logs section, and another that you can type, and press enter to send text in. I've drawn up a little diagram of how I want to make it: (I know this looks awful, it was made in mspaint)
I've gotten the console input part working by using the readline module, but I don't even know where to really start with designing the terminal how I want it, setting text in certain sections, etc. I've looked around at things like terminal-kit, and clci, but either they didn't seem like what I wanted, or their docs/examples were a mess.
I would prefer to do this with node only, not using another application in another language, as all of the stuff going to the console sections will be from the same node application.
I found the blessed library at https://github.com/chjj/blessed. It is based on the ncurses library (written in C, https://en.wikipedia.org/wiki/Ncurses) and it allows you to create different sections with specified heights, widths, etc. in different areas of the terminal. Very useful in theory, and you can follow the advice on the github page, clone the repo into your desktop, look into the test folder, and then run the different files ie. node test/widget-form.js to see the different types of interfaces that you can create in your terminal.
It hasn't really worked for me, because it keeps crashing on my end, but I see that there are a lot of open pull requests and that people are still trying, so it might be working for some, although I think that the usage for some "widgets" is limited. The next best thing I can recommend is either blessed-contrib or neo-blessed, the former being developed by some dude from Facebook. Blessed-contrib is newer so you might have more luck with it, but it's essentially made for visual output, so you'd be able to create a section for a log, a selection menu, a paragraph section, etc. but nothing to create an area where you can input text (so 4/5 of what you need), based on my reading of the documentation, which you'll find here https://github.com/yaronn/blessed-contrib.
I personally think it's dope. Good luck with it, I'm going through the motions of understanding the documentation better myself, and I've delved pretty deeply in it, so feel free to reach out if you need any help.
On another note, let me know if you've found anything else other than the resources I mentioned, which would be very helpful for me, as the alternative right now is for me to code the text user interface myself using C and the ncurses library, which I'm trying to avoid if it's unnecessary :)
EDIT:
I found something for you and for me that might send us in the right direction. Good news. This is the gitter-cli. It uses the library I mentioned above blessed. If you clone the gitter-cli's repo, follow the instructions to get a token and create an account, and join one of the rooms (you can find the name of the rooms like 'gitterHQ/javascript' on the gitter.im website), you'll see that the chat works. There might be some optimizations I'm not aware of, but I recommend just delving into their code while also delving into the blessed documentation to understand how it works. It should give you an idea of what to do next.
It took me a lot of hours to find all these different resources and connect the dots so you should definitely check them out.
I am looking for a way to take a model of a building and allow people to walk through it like a video engine.
We are also looking to run this on a viz wall, which requires OpenGL on Linux and be open source. But Something running on windows or closed source on Linux would be better than nothing.
I have found Panda3D, but I am not sure that will perform well enough for such a large model, the .egg file was over 200MB and took over 8GB of RAM to convert to their binary format.
None of our prefessors know about this, and we are having trouble finding the tools we need.
Try Flux Player, i used this on a school project a while ago.
If i recall correctly i had to export the sketchup file with the extention that the program needs, but i don't remember at the moment which one was that.
http://www.redorbit.com/news/technology/854384/media_machines_releases_flux_studiotm_20_and_flux_playertm_20/
software belongs to "media machines"
this is basically a plugin that allows you to navigate the models on a web browser using your cursor or mouse.
I believe that there are other solutions out there that allow you to do exactly what you want.
I want to build a program that will (as part of what it's doing) display lines organically growing and interacting horizontally across the screen. Here's a sample image, just imagine the lines sprouting from the left and growing to the right:
The lines would look like the lines used on Google Maps Transit Overlay or OnNYTurf's transit pages.
It's a personal project, so I'm open to just about any language and library combination. But I don't know where to start. What have you used in the past to create graphics that are similar to this? What would you recommend? I want it to run on Windows without any extras needed (.Net is fine), and it doesn't have to run elsewhere. I needs to run as an actual program, not javascript in the browser.
There's obviously no 'right' answer to this, but the purpose isn't to start an argument about X better than Y but rather just find a list of graphics toolkits that do simple 2D graphics that people recommend because of their ease of use or community or whatever.
Processing may be just the tool for you.
Like you said, there are many ways to tackle this problem. Me personally, being it is a windows based project, I would go with the .NET based implementation utilizing WPF. There are tutorials on how to use the 2D drawing feature out there ( http://www.wpftutorial.net/DrawOnPhysicalDevicePixels.html for one ) Again, there is no right answer here. I might also pick some new technology and let your project be a mechanism to learn something new, providing you do not have a looming deadline.
BACKGROUND
I am using a commercial application on windows that creates a drawing
This application allows only two output options: (1) save as a bitmap file and (2) print to a printer
the bitmap is useless for my purposes - I want the vectors
Looking at the print output (I sent to the Windows XPS print driver) it seems clear based on the amount of zooming I can do without loss of detail that the underlying vectors are being send to the print driver
Once I get the vectors, I will be writing some code to transform them for some other use.
MY QUESTION
Whart are my options for geting the vectors from the print? (am open to both commercial and open source)
OPTIONS I HAVE THOUGHT OF SO FAR
Take the bitmap and use a program like VectorMagick to. I have tried this approach. It does not produce the fidelity I seek even when the original bitmap is large. Practically speaking I believe that using any tracing approach will not give me the quality vectors I need.
Print to the Adobe PDF driver. This technically works. I have Adobe CS4 so I can print to it save the resulting PDF and then import the PDF into Illustrator and then export as some other vector format. The problem with this approach is money/licensing. I own a personal copy of Adobe CS4 - so this is fine for me. But I need to capture the vectors at work for business purposes - and no I'm not going to install my personal copy of CS4 at work.
Is there a "print driver" that captures the print output directly into a vector format? I have seen some commercial ones via google. If you've used them, I would like to hear about your experience with this technique. I could write my own and in that case do you have links to any existing code that I can start with.
If this is an ongoing solution you need then you might need to buy something or build your own. If it's a onetime affair you might look to use an 'older' Lexmark PCL printer driver. I'd recommend something like the T610. If you download the PCL driver and install it you can modify the defaults and change the Graphics option from XL or Autoselect to GL/2. This will force the driver to output GL/2 output which is vector (GL/2 is a plotter language). This might do the trick for you. Other printer drivers may have the abiltiy to force GL/2 (vs. Raster) but I'm not sure. I use to work for Lexmark and have used this before for a similar requirement.
Ensure you use the Lexmark 'Custom' driver as I don't think the Microsoft-based one support this feature.
...pausing while I investigate a few things............I'm back...
Another option is to find another GL/2 driver or build you own...I just took a few minutes to search the web and came up with a few other options that might work.
Build you own:
I've built drivers (minidrivers) using the Windows Driver Development Kit (DDK), it's quite simple to construct basic drivers. Looks like there is a setting you can set to enable GL/2 output: Enabling HP-GL/2 Vector Graphics Support (PCL-5e) in the GPD
Alternate drivers:
Depending on the OS you are on there is probably a 'generic' GL/2 driver built in. I believe XP has a Hewlett-Packard HP-GL/2 Plotter. You might need to check the license (as with the Lexmark solution) but it might work for you and as it's part of the OS there shouldn't be concern about using it. It's probably written and copyrighted to Microsoft
Keep in mind you will have to do some work to convert GL/2 to whatever output you want but it should be a matter of an simple translator to convert each set of commands. There may be tools out there to help. Here is a quick link to Lexmark GL/2 reference which might be enough to get you going, check out the GL/2 information under the PCL section: Lexmark Technical Reference Guide
Postscript:
The last option I have is to use a generic Postscript driver. Postscript should output the vector images as vector graphics in the Postscript but my knowledge of this is limited at best.
Output:
If you need the output to route to file you can set the port to FILE: which requries user intervention, or install something like Redmon (or connect with me and I'll send you our port monitor that allows for automatic output to file).
Hope this helps in some way.
My favorite is the open source (GPL) PDFCreator
http://sourceforge.net/projects/emfprinter/
Let me explain what I mean by "two-dimensional code editor": imagine of using Inkscape or Gimp in a big canvas (say infinite). The "T - add text" tool is used to write the code. Additionally, all function definitions will be framed and links will connect the called functions.
In other words: you have a very large sheet of (virtual) paper where you can write.
It would be really useful. I don't want to write code as a long list of lines, especially now that big monitors are cheaper.
Is such a code editor out there?
What's your opinion? Would you use a 2d code editor?
I've written 3 or 4 visual editors and my second one worked like this, that was for java and c++ (never published, though I did use it for some published research work)
I still don't like much to write my code 'as a long list of lines'. My point is, after trying a system like this, I tried a windowed system (class outlines in windows, right click to open code editors), then a tree based system...
in the long run (I wrote several apps using all of those), the tree based system with non overlapping windows felt at once most scalable (to different monitor sizes) and foremost, most productive, because dragging the text boxes and links and/or windows in the first version was necessary, without adding much to the programming experience, so it felt wasteful.
If you want to try some of this stuff out, you can google antegram for java (java only) antegram for web (javascript/php/actionscript) and ee-ide (on oogtech.org). I'm not sure if I could dig up the original c++/java textbox + links editor (which could collapse graphs as well, and had an infinite canvas, so pretty close to what you describe).
I'm not working on this as much as I used to as few programmers ever seemed to like it except me, but if you like working the tree way, or feel like adding stuff for your own purposes, ee-ide would be the way to go, as it's nicely modular and easy to extend compared to the rest.
On the commercial side, you can configure visual studio to work with UML-like diagrams. I have a feel it might be a little too heavy (although it's definitely more coding than UML oriented), but I'm not sure, I haven't really tried yet.
This probably doesn't answer your question exactly, but anyway.
Have a look at the NodeBox beta . It is a visual programming environment mostly for creating generative graphics. You can program and edit the nodes with python code, connect and reuse them in multiple ways. (Windows and Mac OS)
Also worth mentioning (in terms of concept) is Field . It is for programming performances and arranges bits of code on a stage/timeline. Very interesting but also very confusing. (Mac OS only)
Third one is vvvv. It is used a lot by graphical artists to create realtime 3d visuals. Node based. (Windows only)
NodeBox and Field are open-source, so if you are looking to create something yourself you can see how it's done there.
Check this out. I came across it today and remembered this question.
Code Bubbles
Developers spend significant time
reading and navigating code fragments
spread across multiple locations. The
file-based nature of contemporary IDEs
makes it prohibitively difficult to
create and maintain a simultaneous
view of such fragments. We propose a
novel user interface metaphor for code
understanding and maintanence based on
collections of lightweight, editable
fragments called bubbles, which form
concurrently visible working sets.
The essential goal of this project is
to make it easier for developers to
see many fragments of code (or other
information) at once without having to
navigate back and forth. Each of these
fragments is shown in a bubble.
A bubble is a fully editable and
interactive view of a fragment such as
a method or collection of member
variables. Bubbles, in contrast to
windows, have minimal border
decoration, avoid clipping their
contents by using automatic code
reflow and elision, and do not overlap
but instead push each other out of the
way. Bubbles exist in a large,
pannable 2-D virtual space where a
cluster of bubbles comprises a
concurrently visible working set.
Bubbles support a lightweight grouping
mechanism, and further support
connections between them.
A quantiative user study indicates
that Code Bubbles increased
performance significantly for two
controlled code understanding tasks. A
qualitative user study with 23
professional developers indicates
substantial interest and enthusiasm
for the approach, despite the radical
departure from what developers are
used to.
http://www.cs.brown.edu/people/acb/codebubbles_site.htm
At one point, LabView had a programming mode like this. You connected program blocks together in a graphical way.
It's been so long since I've used LabView that I don't know if it is still the same.
For me, the MVVM pattern means that there's no code behind the UI controls anyway. The logic is all in a class with properties.
The properties use WPF databinding to update the UI controls. For example, on the form or window, page, whatever, MySearchButton.IsEnabled is bound to ViewModel.MySearchButtonIsEnabled property. So the app logic runs in the ViewModel class and just sets its own properties and the UI updates automatically.
Although this is specific to MS WPF the pattern actually stems from SmallTalk and is found across the development field as MVP. Without WPF one would need to write the databinding or 'presenter' logic, which is common.
This means the UI can be torn off and a new one pasted-in really quickly and with little code knowledge from the UI guy - who, in an ideal world, is a crack creative guy that drives a 70s Citroen.
So my point is that, although it sounds like a neat innovation, a 2D editor like this would be assisting a coding style that is no longer considered optimal.