Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 5 years ago.
Improve this question
I am researching into using TensorFlow and am trying to decide which language to write in.
I am currently using Clojure in my day job (I know you can use Java so should I check out a Clojure wrapper). I have also started to learn Haskell. As TensorFlow is very mathematically based maybe Haskell would be the best language to write it in.
I have read that TensorFlow is mainly written in C++ and so Python is the main language that people use.
Is it best to write TensorFlow in Python or does Clojure/Java or Haskell work just as well?
It depends on what your use case is.
For exploring machine learning concepts or for doing machine learning research Python is your best bet. It has the most features, documentation and user base to help with support.
Java and some other language bindings are geared more towards integrating models with existing code (into production pipelines, apps, your-other-production servers etc.) and don't have the convenience of breadth of API that Python provides (quoting from the api documentation - "The Python API is at present the most complete and the easiest to use, but other language APIs may be easier to integrate into projects and may offer some performance advantages in graph execution.")
The Haskell bindings might be interesting, but again, if you're exploring TensorFlow, you're probably better off with Python to begin with. If and when you need to integrate models into other software, you will be able to export them from Python and import/execute them in other languages.
Hope that helps.
Related
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 1 year ago.
Improve this question
We are trying to train an LSTM under anaconda (spyder) from a dataset of size 333113kb. (3628801 rows * 31columns) The data is stored in a .csv file and is imported using the pandas library. The execution is too slow and somtimes spyder crashes.
NB : we are using an Intel Core (TM) i5-8300H CPU 2.3 GHz with 8Go of RAM.
not directly an IDE but I like to use Visual Studio Code because it's an excellent way to go even for machine learning and data science.
it works on any OS
supports many Technologies besides Python, C#, JS etc.
open source and lightweight
VS Code is integrated with PyLint
You can perform unit testing on your machine learning models easily
for me VS Code makes working with SQL, .NET, Node.js and many other Tools a lot easier. It's a great code editor that supports you with operations like debugging, task running, version control and many other things that a full featured IDE can also do.
Is this like university-wise project or commercial? You could try Google Colab.
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 5 years ago.
Improve this question
Recently I decided to learn MXNet, as some code I need to use, is written using this API.
However, I would like to know which are the advantages and disadvantages of MXNet compared to the other Deep Learning Libraries out there.
Perhaps the biggest reason for considering MXNet is its high-performance imperative API. This is one of the most important advantages of MXNet to other platforms. Imperative API with autograd makes it much easier and more intuitive to compose and debug a network. PyTorch also supports imperative API, but MXNet is the only platform AFAIK that supports hybridization, which effectively allows your imperative model to be converted to a symbol for similar performance to symbolic API. Here is a link to tutorials on Gluon, MXNet's imperative API: http://gluon.mxnet.io/
Given that you're using an example code, it is possible that the example was written using symbolic API. You may notice MXNet's advantage in symbolic API when training on many GPUs. Otherwise you won't notice much of a difference (except perhaps in some memory usage).
Tensorflow does have a one year head-start to MXNet and as a result it has a larger user base, but it only supports symbolic API (imperative API is very new and is only meant for experimentation), which is significantly harder to debug a network when you run into issues. However MXNet has quickly caught up in features and with 1.0 release, I don't think there is anything in TF that MXNet doesn't support.
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 6 years ago.
Improve this question
I wanted to ask little bit generic question about Java API's. I'm new in Java, learning myself and of course I enjoy it while learning. But when it comes to Java API's, to me it's hard to understand even if I know OOP's Concepts. Actually I often confuse while reading JavaDoc's. Agree that there are bunch of information about what class's(interface, enum..) has, what they extend, implements or whatever. Even though I couldn't use them without looking from the internet.
So, what is the best way of learning those API's in general, just give me your followed way while learning Java.
The best way to learn Java is to read articles with examples. It is common approach for newbies and professionals. JavaDocs is dry, too formal and really are often confusing (especially for methods from Stream API with 3+ generic parameters).
The way I did was to read certification books, for OCA and OCP, they start from the very basic, and builds up on the top of that, with examples and exercises.
It is a really nice way to build your core knowledge of the language. With that solidified knowledge you can start reading articles about different subjects in the Java universe ...
Just be patient, code as much as possible, and it will become natural aftar some time.
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 5 years ago.
Improve this question
the project I am working on ship a package that contains API for different languages: Java, Python, C#, and others. All these APIs shared mostly the same documentation. The documentation should be available in PDF and HTML separately on our website. The user usually download/browse the one it is interested in.
Currently we use sdocml, but we are not that satisfied and so we want to move to a more up to date tool and we are considering Sphinx.
Looking at the Sphinx documentation I cannot clearly figure out how:
1- say to generate the docs for a certain API (for instance the Java one)
2- does autodoc works for any domain?
3- is there a c# extension?
Any help is most welcome!
The best way to combine different languages in one Sphinx project is to write docs without autodoc or other means of automatic generation. For the most part they are available only for Python and even if some extension out there does allow other languages, you will be buried under different workflows before you even notice.
Salvage your docs from the code and write them in concise manner in a separate docs folder of your project or even separate repository. You could use the generic Sphinx directories like class or method with no attachment to the code and for virtually any major programming language. I for myself did a project like that, where I needed to combine C, C++ and Python code in one API and it was done manually.
If you create this kind of detached project, the maintenance should be much of an issue. It's not much harder, than autodoc workflow. What for PDF and HTML - any Sphinx project allows that. See their docs for details on different builders like latexpdf or html.
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 6 years ago.
Improve this question
I have a large legacy .NET application that has evolved and grown over the years to include many components and moving parts. I want to develop a strategy for developing automated unit and integration tests for this application and to that end I think a graphical representation would be key.
What I am picturing is some sort of diagram I could use to guide the process of writing up the test cases, help achieve better coverage, and eventually refer back to once a specific test fails. Does anyone have any thoughts on what type of diagram could fulfill this goal? My guess is this would be a variant of the classic functional block diagram, but I have not found examples that specifically relate to the design of an automated testing strategy.
Could this be what you are looking for?
The UTP provides extensions to UML to
support the design, visualization,
specification, analysis, construction,
and documentation of the artifacts
involved in testing. It is
independent of implementation
languages and technologies, and can be
applied in a variety of domains of
development.
UML Testing Profile: http://utp.omg.org/