...

Then we started to enhance the distributions to make sure that we can plug other distribution for having controlled random number.

Ok, did you commit the code ?

No I lost my internet connection at home ;( Ask the company to check.

...

- Do a public call to know what is missing for you: this is this mail :) Didier would like to work on concrete cases. I love that attitude So tell us :)

Hernan ?? Alex: ?? pvalue? better distribution? Vincent: covariance? CPA? Philippe: times series Serge R frames? Sami: Better random number and various distributions?

- Organise a two day lectures with practices on concrete case in September with a recording session. Either at IRD Bondy or Lille.

- Make sure that the Numerical Method book will get on lulu :) with a better cover and title :)

Thank you Stef for pushing ! Can we move this discussion of the SciSmalltalk mailing-list ?

On Wed, Mar 2, 2016 at 11:39 PM, Alexandre Bergel <alexandre.bergel@me.com> wrote:

Hi !

...

- Do a public call to know what is missing for you: this is this mail :) Didier would like to work on concrete cases. I love that attitude So tell us :)

This is an excellent initiative! I personally need the following: - Shapiro-Wilk test to check for data normality - Kruskal-Wallis to analyze differences in median values between samples - Bonferroni-Dunn to support multiple comparison and controlling the group error rate - Mann-Whitney U test is a nonparametric statistical test that operates over two samples of the same size.

Great list ! At the moment, there is only Kolmogorov-Smirnov test implemented. Can you create issues in the SciSmalltalk issue tracker in order to keep track of your needs ? The issue tracker is here: https://github.com/SergeStinckwich/SciSmalltalk/issues You can also join the SciSmalltalk mailing-list: https://groups.google.com/forum/#!forum/scismalltalk Thank you Alex. -- Serge Stinckwich UCBN & UMI UMMISCO 209 (IRD/UPMC) Every DSL ends up being Smalltalk http://www.doesnotunderstand.org/

Thanks for this great list!!!

Alex what would be cool is to sketch a test with some data some that 
like that it is a good incentive for didier. :)

> Hi !
>
>>    - Do a public call to know what is missing for you: this is this mail :)
>>     Didier would like to work on concrete cases. I love that attitude
>>     So tell us :)
> This is an excellent initiative!
> I personally need the following:
> 	- Shapiro-Wilk test to check for data normality
> 	- Kruskal-Wallis to analyze differences in median values between samples
> 	- Bonferroni-Dunn to support multiple comparison and controlling the group error rate
> 	- Mann-Whitney U test is a nonparametric statistical test that operates over two samples of the same size.
>
> Cheers,
> Alexandre
>

On Thu, Mar 3, 2016 at 6:25 AM, Hernán Morales Durand <hernan.morales@gmail.com> wrote:

Hi Stef,

For bioinformatics research it would be **really** nice to have:

- Tree creation algorithms: UPGMA, WPGMA. - MCMC methods: Metropolis-adjusted Langevin algorithm (MALA and SMMALA), Metropolis-coupled MCMC (MC3) would enable Bayesian inferences and to work with evolution of genes. I cannot tell how important is to have such algorithms implemented in Smalltalk. - Mahalanobis distance for analyzing bacterial plasmids. - Random Forest is used a lot for gene expression, protein-protein interaction predictions, genome-wide association studies, etc . For the argument "let's use FFI for it" please remember many of these methods are extended, like enriched random forest, for research with specialized data.

Can you add enhancement issues in the SciSmalltalk issue tracker : https://github.com/SergeStinckwich/SciSmalltalk/issues in order to keep them in mind ? We are trying to create a pure Smalltalk library, so we try to avoid FFI as much as possible :-)

There are implementations of naive Bayes classifier in NaturalSmalltalk (which could be adapted to SciPharo) http://smalltalkhub.com/#!/~hernan/NaturalSmalltalk

Great, if we can move them to SciSmalltalk.

As a side note I found an interesting history report of SciPy. From the slide 9 onwards:

http://sssslide.com/www.slideshare.net/shoheihido/sci-pyhistory

Presentation shows SciPy history (credited contributors with pictures, very nice!). As you can se they even have a Foundation NumFOCUS for raising money.

For sure, money always help ;-) The only problem at the moment is that the Pharo community is much smaller at the moment than the Python one. Thank you for your help. -- Serge Stinckwich UCBN & UMI UMMISCO 209 (IRD/UPMC) Every DSL ends up being Smalltalk http://www.doesnotunderstand.org/

-----Message d'origine----- De : Pharo-dev [mailto:pharo-dev-bounces@lists.pharo.org] De la part de Serge Stinckwich Envoyé : jeudi 3 mars 2016 11:04 À : Any question about pharo is welcome Cc : Didier Besset; Souissi Sami; Pharo Development List Objet : Re: [Pharo-dev] [Pharo-users] Call about Numerical Methods in Pharo :)

On Thu, Mar 3, 2016 at 10:43 AM, Blondeau Vincent <vincent.blondeau@worldline.com> wrote:

...

Hi,

That's a very good idea. For now, I use a Pharo-R binding to make the statistics, but it is better if it is directly in the image.

For my part, it should be nice to have: - Principal Component Analysis - Time series

I already a very simple time series class. What kind of operations you want to do on time series ?

For now, I am not using time series. But one year ago, I was using Holt Winters and Seasonal Decomposition (HoltWinters and stl functions under R). And it is possible that I reuse them in a close future.

...

- Correlation with several variables (Correlation matrix) + significance tests (pearson, spearman...) - Standard statistical tests: normality, variance, mean, distribution - And the more important is to have nice tools to visualize the data (Box plot, histogram, scatterplots, PCA ones) with a simple API for standard things but that we can customize. The best is that we are able to draw regression lines on them. But I think that is more a Roassal part.

You have to ask Alex to the viz :-)

;)

Please join the SciSmalltalk ml in order to express your needs ;-) -- Serge Stinckwich UCBN & UMI UMMISCO 209 (IRD/UPMC) Every DSL ends up being Smalltalk http://www.doesnotunderstand.org/

!!!************************************************************************************* "Ce message et les pièces jointes sont confidentiels et réservés à l'usage exclusif de ses destinataires. Il peut également être protégé par le secret professionnel. Si vous recevez ce message par erreur, merci d'en avertir immédiatement l'expéditeur et de le détruire. L'intégrité du message ne pouvant être assurée sur Internet, la responsabilité de Worldline ne pourra être recherchée quant au contenu de ce message. Bien que les meilleurs efforts soient faits pour maintenir cette transmission exempte de tout virus, l'expéditeur ne donne aucune garantie à cet égard et sa responsabilité ne saurait être recherchée pour tout dommage résultant d'un virus transmis. This e-mail and the documents attached are confidential and intended solely for the addressee; it may also be privileged. If you receive this e-mail in error, please notify the sender immediately and destroy it. As its integrity cannot be secured on the Internet, the Worldline liability cannot be triggered for the message content. Although the sender endeavours to maintain a computer virus-free network, the sender does not warrant that this transmission is virus-free and will not be liable for any damages resulting from any virus transmitted.!!!"

Hi Vincent, it would perhaps be helpful if you could specify your proposal "Standard statistical tests: normality, variance, mean, distribution" a bit, because eventually some of these are already implemented and would just need a wrapper to be easily usable. perhaps it would be enough if one would know which R-functions you use or so, and what kind of return you'd expect. werner On 03/03/2016 10:43 AM, Blondeau Vincent wrote:

Hi,

That's a very good idea. For now, I use a Pharo-R binding to make the statistics, but it is better if it is directly in the image.

For my part, it should be nice to have: - Principal Component Analysis - Time series - Correlation with several variables (Correlation matrix) + significance tests (pearson, spearman...) - Standard statistical tests: normality, variance, mean, distribution - And the more important is to have nice tools to visualize the data (Box plot, histogram, scatterplots, PCA ones) with a simple API for standard things but that we can customize. The best is that we are able to draw regression lines on them. But I think that is more a Roassal part.

Thanks for the initiative,

Cheers, Vincent

...

-----Message d'origine----- De : Pharo-users [mailto:pharo-users-bounces@lists.pharo.org] De la part de stepharo Envoyé : mercredi 2 mars 2016 14:17 À : Pharo Development List; Any question about pharo is welcome; Didier Besset; Souissi Sami Objet : [Pharo-users] Call about Numerical Methods in Pharo :)

Hi guys

I met Didier Besset and we had a great hacking session and discussions with Serge Stinckwich. Didier would like to help Pharo and the numerical part of it. ***Big thanks*** Didier.

We would like to do several things:

- Work on "Hows to" The numerical methods in Pharo is good but the gap between us and the math is too large :) so the idea is to have a series of "how to ..." - histomgram (simple, based on distribution)

- Improve the SciPharo/NumPha (previously SciSmalltalk) library This morning we started to implement a ComponentPrincipalDecomposition by combining two of the objects available in SciPharo. Then we started to enhance the distributions to make sure that we can plug other distribution for having controlled random number.

- Do a public call to know what is missing for you: this is this mail :) Didier would like to work on concrete cases. I love that attitude So tell us :)

Hernan ?? Alex: ?? pvalue? better distribution? Vincent: covariance? CPA? Philippe: times series Serge R frames? Sami: Better random number and various distributions?

- Organise a two day lectures with practices on concrete case in September with a recording session. Either at IRD Bondy or Lille.

- Make sure that the Numerical Method book will get on lulu :) with a better cover and title :)

Stef

!!!************************************************************************************* "Ce message et les pièces jointes sont confidentiels et réservés à l'usage exclusif de ses destinataires. Il peut également être protégé par le secret professionnel. Si vous recevez ce message par erreur, merci d'en avertir immédiatement l'expéditeur et de le détruire. L'intégrité du message ne pouvant être assurée sur Internet, la responsabilité de Worldline ne pourra être recherchée quant au contenu de ce message. Bien que les meilleurs efforts soient faits pour maintenir cette transmission exempte de tout virus, l'expéditeur ne donne aucune garantie à cet égard et sa responsabilité ne saurait être recherchée pour tout dommage résultant d'un virus transmis.

This e-mail and the documents attached are confidential and intended solely for the addressee; it may also be privileged. If you receive this e-mail in error, please notify the sender immediately and destroy it. As its integrity cannot be secured on the Internet, the Worldline liability cannot be triggered for the message content. Although the sender endeavours to maintain a computer virus-free network, the sender does not warrant that this transmission is virus-free and will not be liable for any damages resulting from any virus transmitted.!!!"

Hi Alex, This binding allows to execute (almost) any function provided by R via Native Calls. You can find some documentation here: http://smalltalkhub.com/#!/~VincentBlondeau/RProjectConnector I just released a new stable version, you should be able to load it from the catalog (Pharo40 only). However, you must install R and copy the libraries in the VM folder (see doc). And it is working only with pre-spur images (the migration to FFI has to be done). Try to run the tests (in RConnector package). Iff they are green, you succeeded to configure the binding. And you can plot the famous Iris Data: iris := 'eval' asREval: { RObject findVar: 'iris' }. 'plot' asREval: {(iris at: 'Petal.Length'). (iris at: 'Petal.Width'). (#pch -> 21). (#xlab -> 'length'). (#ylab -> 'Width'). (#bg ->((iris at: 'Species') collect: [ :value | {'red'. 'green3'. 'blue'} at: value ])). (#main -> 'Edgar Anderson''s Iris Data')}. plot(iris$Petal.Length, iris$Petal.Width, pch=21, bg=c("red","green3","blue")[unclass(iris$Species)], main="Edgar Anderson's Iris Data") is the genuine R function. And inspect them: iris inspect You can use any collection of pharo and put it as argument, it should do the boxing automatically. Warning: while testing new functions, don’t forget to save because evaluation of a non-existing R function throws a error that is not caught and makes Pharo crashing. Cheers, Vincent De : Pharo-dev [mailto:pharo-dev-bounces@lists.pharo.org] De la part de Alexandre Bergel Envoyé : mardi 8 mars 2016 20:23 À : Any question about pharo is welcome Cc : Didier Besset; Souissi Sami; Pharo Development List Objet : Re: [Pharo-dev] [Pharo-users] Call about Numerical Methods in Pharo :) Vincent, can you tell a bit more what this binding is about? how can I try? Alexandre -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. On Mar 3, 2016, at 6:43 AM, Blondeau Vincent <vincent.blondeau@worldline.com<mailto:vincent.blondeau@worldline.com>> wrote: Hi, That's a very good idea. For now, I use a Pharo-R binding to make the statistics, but it is better if it is directly in the image. For my part, it should be nice to have: - Principal Component Analysis - Time series - Correlation with several variables (Correlation matrix) + significance tests (pearson, spearman...) - Standard statistical tests: normality, variance, mean, distribution - And the more important is to have nice tools to visualize the data (Box plot, histogram, scatterplots, PCA ones) with a simple API for standard things but that we can customize. The best is that we are able to draw regression lines on them. But I think that is more a Roassal part. Thanks for the initiative, Cheers, Vincent -----Message d'origine----- De : Pharo-users [mailto:pharo-users-bounces@lists.pharo.org] De la part de stepharo Envoyé : mercredi 2 mars 2016 14:17 À : Pharo Development List; Any question about pharo is welcome; Didier Besset; Souissi Sami Objet : [Pharo-users] Call about Numerical Methods in Pharo :) Hi guys I met Didier Besset and we had a great hacking session and discussions with Serge Stinckwich. Didier would like to help Pharo and the numerical part of it. ***Big thanks*** Didier. We would like to do several things: - Work on "Hows to" The numerical methods in Pharo is good but the gap between us and the math is too large :) so the idea is to have a series of "how to ..." - histomgram (simple, based on distribution) - Improve the SciPharo/NumPha (previously SciSmalltalk) library This morning we started to implement a ComponentPrincipalDecomposition by combining two of the objects available in SciPharo. Then we started to enhance the distributions to make sure that we can plug other distribution for having controlled random number. - Do a public call to know what is missing for you: this is this mail :) Didier would like to work on concrete cases. I love that attitude So tell us :) Hernan ?? Alex: ?? pvalue? better distribution? Vincent: covariance? CPA? Philippe: times series Serge R frames? Sami: Better random number and various distributions? - Organise a two day lectures with practices on concrete case in September with a recording session. Either at IRD Bondy or Lille. - Make sure that the Numerical Method book will get on lulu :) with a better cover and title :) Stef !!!************************************************************************************* "Ce message et les pièces jointes sont confidentiels et réservés à l'usage exclusif de ses destinataires. Il peut également être protégé par le secret professionnel. Si vous recevez ce message par erreur, merci d'en avertir immédiatement l'expéditeur et de le détruire. L'intégrité du message ne pouvant être assurée sur Internet, la responsabilité de Worldline ne pourra être recherchée quant au contenu de ce message. Bien que les meilleurs efforts soient faits pour maintenir cette transmission exempte de tout virus, l'expéditeur ne donne aucune garantie à cet égard et sa responsabilité ne saurait être recherchée pour tout dommage résultant d'un virus transmis. This e-mail and the documents attached are confidential and intended solely for the addressee; it may also be privileged. If you receive this e-mail in error, please notify the sender immediately and destroy it. As its integrity cannot be secured on the Internet, the Worldline liability cannot be triggered for the message content. Although the sender endeavours to maintain a computer virus-free network, the sender does not warrant that this transmission is virus-free and will not be liable for any damages resulting from any virus transmitted.!!!" !!!************************************************************************************* "Ce message et les pièces jointes sont confidentiels et réservés à l'usage exclusif de ses destinataires. Il peut également être protégé par le secret professionnel. Si vous recevez ce message par erreur, merci d'en avertir immédiatement l'expéditeur et de le détruire. L'intégrité du message ne pouvant être assurée sur Internet, la responsabilité de Worldline ne pourra être recherchée quant au contenu de ce message. Bien que les meilleurs efforts soient faits pour maintenir cette transmission exempte de tout virus, l'expéditeur ne donne aucune garantie à cet égard et sa responsabilité ne saurait être recherchée pour tout dommage résultant d'un virus transmis. This e-mail and the documents attached are confidential and intended solely for the addressee; it may also be privileged. If you receive this e-mail in error, please notify the sender immediately and destroy it. As its integrity cannot be secured on the Internet, the Worldline liability cannot be triggered for the message content. Although the sender endeavours to maintain a computer virus-free network, the sender does not warrant that this transmission is virus-free and will not be liable for any damages resulting from any virus transmitted.!!!"