Science topic

Scala - Science topic

Explore the latest questions and answers in Scala, and find Scala experts.
Questions related to Scala
  • asked a question related to Scala
Question
1 answer
Cerco la traduzione italiana dalla scala Organizational Justice di Colquitt, 2001.
In realtà mi servirebbe solo capire come avete tradotto "outcome" e "candid in communications", tutto il resto è chiaro.
grazie
p.s.
A meno che non sia stata fatta una validazione linguistica alla quale doversi attenere.
Relevant answer
Answer
solved. thanks to all.
  • asked a question related to Scala
Question
32 answers
Which programming language do you need for proper data processing?
R
Python
Scala
Julia
Relevant answer
Answer
Dear Mohsen Ghorbian,
Java , C#, Matlab
Regards,
Shafagat
  • asked a question related to Scala
Question
9 answers
I’m looking for a mathematical approach to identify the distribution of my data. Due to the fact that I need this approach in Scala, I cannot use the R packages fitdistr or fitdistrplus and of course I cannot use the R plots, that’s why I need a mathematical approach.
I’m looking for a mathematical approach to identify the distribution of my data. Due to the fact that I need this approach in Scala, I cannot use the R packages fitdistr or fitdistrplus and of course I cannot use the R plots, that’s why I need a mathematical approach.
I tried to use Sparks Statistics.kolmogorovSmirnovTest on my data in order to test for Normal Distribution and the p-Value was extremely bad. But in the following histogram we can see that the data is kind-of normal distributed. I assumed that if the test works badly for Normal Distribution there is no need in testing for other types of distribution.
I’ve read different posts across the internet suggesting that the Kolmogorov Smirnov Test is not really reliable and that the Shapiro-Wilk Test does not work well with large data.
I’ve also took a look at MLE but somehow I can’t really understand how MLE can help me determine if my data follows for instance a Normal Distribution. I can't see how estimating mu and sigmacan help me. Is there something I’m missing?
Searching on the internet for this topic, I found only posts in which the distribution of the data was identified with the help of different tools (R or Matlab)
Relevant answer
Answer
If Scala has kernel density estimation functionality then maybe that will solve your problem, https://en.wikipedia.org/wiki/Kernel_density_estimation.
Otherwise I tend to agree with Desmond, Pearson's goodness of fit. The negatively skewed distribution shown in your histogram could be modelled by a beta or gamma distribution, you'll have to vary the parameters. Probably best to do this by eye initially.
  • asked a question related to Scala
Question
4 answers
How to connect SQL database server using Scala to exchange data during Resource Scheduling for execution of workloads?
Relevant answer
Answer
Apply ODBC connection with a DSN.
  • asked a question related to Scala
Question
1 answer
Compute listed vertex-based similarity measures for all the pairs of nodes in label data file(Node1, Node2, status - weak/strong) . These similarity measures are computed between two nodes by utilizing neighborhood and/or node information of both nodes.
Common neighbors
Jaccard coefficient
Adamic/Adar
Preferential Attachment
Any scala library can solve this problem??
Relevant answer
Answer
As simple as you can think just use scala to code it and solve it, it may be Difficult a little but you will understand the problem better.
hope you all the best.
  • asked a question related to Scala
Question
2 answers
I'm now studying about unbalanced text classification.
Most of the methods are obtained in weka.
However I'd like to classify the big text data on Spark. I can't find the scala SMOTE for text classification. Could you please help me with any scala code for unbalanced text data classification?
Relevant answer
Answer
  • asked a question related to Scala
Question
6 answers
Hello everyone,
I would like to perform machine learning and deep learning in a near futur on clinical data, coming from different sources and technologies (Omics data, Imaging data, and so on).
These data will have to be stored in a structured way in a datawarehouse and will be accessed thanks to a 10Gbit network and I will have some clusters available.
My question is the following : What could be the best architecture to use to run parallel calculations/deep learning/machine learning on these data ? I first though at Spark but would it really be appropriate to wide data (often few observations for lot's of variables - thousands to millions biomarkers analyzed) ?
Is there other framework(s) more adapted to this kind of question ?
What kind of advices would you give me to stick to the best "big data" practice ?
If Spark seems the good answer to my problem, would you use R language (as library sparklyr) or Scala/Python instead to perform deep learning/machine learning ?
Thanks in advance for all your advices,
Best regards
Relevant answer
Answer
Hi Nathalie,
My first advice is to experiment of a small portion of the data and determine the complexity of the prediction task. This will give you a feeling about how much computer power you will really need.
Second, consider R, I have always used R and never looked back. I do know many other platforms can eventually outpace R (particularly w.r.t. speed). Nevertheless, when I go for hardcore modeling I use 2 tricks to improve speed: i) Microsoft R Open (entirely free), which enables full-range parallel computation and ii) the R package Rccp, to run on C++ as much as possible.
Let me know if I can be helpful! Cheers,
F
  • asked a question related to Scala
Question
3 answers
Since Java is included in that range, I'm curious about Scala.
Note that Scala is more difficult to analyze than Java for several reasons, including but not limited to:
- Compile-time code generation (e.g., macros)
- Expressive type system
Relevant answer
Answer
see lots of examples: http://spinroot.com/cobra/papers
  • asked a question related to Scala
Question
3 answers
In the attached article we propose a qualy-quantitative assessment carried out with mix method: evaluation scales and narrative tools (narration, patient agenda, autobiography, therapeutic emplotment....). What do you think? Are there other consolidated experiences in this regard?
Best!!
Giovanan
Relevant answer
Answer
Thank's, Mukesh and Margarita.
  • asked a question related to Scala
Question
9 answers
I study functional programming with Scala, but I have a problem with the recursive algorithms because they are difficult to approach.
Relevant answer
Answer
Another natural place to look at would be Martin Odersky's Coursera course:
Martin Odersky of École Polytechnique Fédérale de Lausanne is the inventor of Scala.I have worked through this course and can recommend it very much.
  • asked a question related to Scala
Question
4 answers
I have 144 diffraction images for a crystal of high mosaicity about 2 and low resolution about 3.14.
the crystal has been prepared and soaked with ligand. I would like to solve the crystal structure looking merely for a substrate binding site.
I tried using IMosflm, however the results of integration showed that all reflections are partial and there is no full reflection at all.
Then I used scala from CCP4 to do 2D integration for partial reflections, then check the density map using molecular replacement (protein has been deposited in PDB) with Phoenix. I can Max have LLG 900 and TFZ of 11. I can only recognize coarse features like helices, however the map is not clear for many side chains as well as at the binding site.
Any suggestion to improve the results using parameters adjustment of iMosflm, Scala, Phoenix MR. OR even new suggested software.
I appreciate any suggestion.
thanks
Relevant answer
Answer
Hi Mohammed,
I totally agree with Jonathan. Partial reflections are not the problem. The problems are probably due to bad quality of diffraction. For the data process, you may also try HKL2000.
Ps. Just a reminder, you may know it, make sure to select a correct space group when you process data, check it with Pointless. Sometimes you can get a MR solution with a low symmetry/accurate space group, but it can't be refined.
Good luck!
  • asked a question related to Scala
Question
1 answer
I am implementing a phylogenetic library in Scala and stumbled over numerical problems with the GTR-model (Yang 1994). The transition probabilities after a given time t of a rate-matrix Q are given by the matrix exponential exp(Qt). According to the literature, this is often computed using the Eigen-Decomposition (V * D * V^-1) of Q. While this decomposition is numerical stable, the exponential (V * exp(Dt) * V^-1) is not. For increasing values of t (e.g. t > 10) the probabilities do not sum to 1.
This problem holds true for standard numerical platforms like octave and several java-linalg-frameworks. In mathematica however, the result is computed correctly. Does mathematica use and arbitrary precision arithmetic?
Does someone experience the same problems or has any idea how to handle it (e.g. rescaling of probabilities to 1)?
Relevant answer
Answer
I have not met the same problem but the following idea should work. You may handly check the matrix D, and making the largest eigenvalue 0. the corresponding eigenvector should be (1,1,1,1).
  • asked a question related to Scala
Question
4 answers
I am working to develop rigid body algorithms using scala. Algorithms will be designed to solve kinematic and dynamic multibody rigid body dynamics.
Relevant answer
Answer
Dear Jawad, "Best" is a relevant term as it relates to many perspectives. If you are familiar with Matlab, you may like Scalalab https://code.google.com/p/scalalab/ and for matrices https://github.com/twitter/algebird