Mailman 3 June 2016 - scikit-learn

how to create and execute a machine learning models in Java/JVM based application (in production) using Python
by Gaurav gupta June 13, 2016

June 13, 2016

Hi All, Could you please guide me on how to *create and execute *a machine learning models/statistical models (regression, Decision tree, K means clustering, Naive bayes, scorecard/linear/logistic regression etc. and GBM, GLM ) in *Java/JVM based application* (in production). We have an ETL sort of Java based product where one can do most of data Preparation steps for machine learning, like data ingestion from JDBC, files, HDFS, No SQL etc., joins and aggregations etc.(which are required for Feature engineering) and now we want to add Analytics capabilities using machine learning/statistical modeling. Right now, we are using JPMML- evaluator <https://github.com/jpmml/jpmml-evaluator> to score the models created in PMML format using R and python (and Knime) but it needs three separate and unconnected steps:- 1- first step for data preparation in our Java/JVM application and save the sampling data (training and test) data in csv file or in DB, - *<JAVA/JVM BASED application>* *2- Create a machine learning Model in R and python (and Knime) and export it in PMML 4.2 format - <NON JAVA BASED >* 3- Import/deploy the PMML in our Java based application and use JPMML evaluator to execute it in production. *<JAVA BASED>* I am sure it's a common problem in machine learning as generally in Production JAVA is preferred over Python or R. Could you suggest what is the better approach(s) to *create* as well as *execute* a python/scikit based machine learning model in JVM based application. What are your thought to achieve the steps # 2 and #3 more seamlessly in a JVM based application, without compromising *performance and usability*:- 1- Call a java program which internally calls the python scikit script <http://stackoverflow.com/questions/12738827/how-can-i-call-scikit-learn-cla…>(under the hood) to create a model in PMML <https://github.com/jpmml/jpmml-sklearn> and then use JPMML evaluator. It will pretend to the user that he is in a single JVM based application (better usability). I am not sure what are the limitations and short coming of using PMML as not all features are supported in jpmml-sklearn <https://github.com/jpmml/jpmml-sklearn>. 2- Call a java program which internally calls the python script and do the model creation as well as execution in an external python environment and serialized the model and the results in a file/csv or in memory DB (or cache, like hazelcast) from where the parent Java application will fetch the results etc.. I researched that I can’t use Jython for executing Sci-kit models <http://stackoverflow.com/questions/12738827/how-can-i-call-scikit-learn-cla…> . 3- Can I use Jep <https://github.com/mrj0/jep> (Embed Python in Java) to embed Cpython in JVM ? Does anybody tried it for sci-kit models? Alternatively, I should explore to use Mahout or weka - java based machine learning libraries in my JVM based application. (I need to support both windows and non-windows platforms) I am also exploring H2Oai which is java based. Does anybody tried it. Regards Gaurav

2 1

EuroSciPy 2016 Call for Papers Extended
by federico vaggi June 8, 2016

June 8, 2016

Hi everyone, The call for contributions (talks, posters, sprints) is still open until June 24th. EuroSciPy 2016 takes place in Erlangen, Germany, from the 23 to the 27 of August and consists of two days of tutorials (beginner and advanced tracks) and two days of conference representing many fields of science, with a focus on Python tools for science. A day of sprints follows (sprints TBA). The keynote speakers are Ga?l Varoquaux (you might have heard of him) and Abby Cabunoc Mayes and we can expect a rich tutorial and scientific program! Videos from previous years are available at https://www.youtube.com/playlist?list=PLYx7XA2nY5GeQCCugyvtnHMVLdhYlrRxH and https://www.youtube.com/playlist?list=PLYx7XA2nY5Gcpabmu61kKcToLz0FapmHu We are particularly eager to receive proposals from newcomers. EuroSciPy is a very welcoming conference, and we are very curious to hear how you use Python/machine learning in your every day research. Visit us, register and submit an abstract on our website! https://www.euroscipy.org/2016/ SciPythonic regards, The EuroSciPy 2016 team

1 0

NB-SVM Implementation
by Ivo Petkantchin June 8, 2016

June 8, 2016

Hello, For a university project I worked on a Sentiment Analysis challenge (Movie Reviews) and implemented a version of NBSVM as described in this paper: http://nlp.stanford.edu/pubs/sidaw12_simple_sentiment.pdf If I am not wrong there is no NBSVM class in scikit-learn. That is why I would like to contribute by coding a NB matrix class if the work is not done by someone else already. Best regards, Ivaylo Petkantchin

2 2

Fitting Lognormal Distribution
by Startup Hire June 6, 2016

June 6, 2016

Hi all, Hope you are doing good. I am working on a project where I need to do the following things: 1. I need to fit a lognormal distribution to a set of values [I know its lognormal by a simple XY scatter plot in excel] 2. I need to find the intersection of the lognormal distribution so that I can decide cut-off values based on that. Can you guide me on (1) and (2) can be achieved in python? Regards, Sanant

8 21

memory efficient feature extraction
by Roman Yurchak June 6, 2016

June 6, 2016

Dear all, I was wondering if somebody could advise on the best way for generating/storing large sparse feature sets that do not fit in memory? In particular, I have the following workflow, Large text dataset -> HashingVectorizer -> Feature set in a sparse CSR array on disk -> Training a classifier -> Predictions where the the generated feature set is too large to fit in RAM, however the classifier training can be done in one step (as it uses only certain rows of the CSR array) and the prediction can be split in several steps, all of which fit in memory. Since the training can be performed in one step, I'm not looking for incremental learning out-of-core approaches and saving features to disk for later processing is definitely useful. For instance, if it was possible to save the output of the HashingVectorizer to a single file on disk (using e.g. joblib.dump) then load this file as a memory map (using e.g. joblib.load(.., mmap_mode='r')) everything would work great. Due to memory constraints this cannot be done directly, and the best case scenario is applying HashingVectorizer on chunks of the dataset, which produces a series of sparse CSR arrays on disk. Then, - concatenation of theses arrays into a single CSR array appears to be non-tivial given the memory constraints (e.g. scipy.sparse.vstack transforms all arrays to COO sparse representation internally). - I was not able to find an abstraction layer that would allow to represent these sparse arrays as a single array. For instance, dask could allow to do this for dense arrays ( http://dask.pydata.org/en/latest/array-stack.html ), however support for sparse arrays is only planned at this point ( https://github.com/dask/dask/issues/174 ). Finally, it is not possible to pre-allocate the full array on disk in advance (and access it as a memory map) because we don't know the number of non-zero elements in the sparse array before running the feature extraction. Of course, it is possible to overcome all these difficulties by using a machine with more memory, but my point is rather to have a memory efficient workflow. I would really appreciate any advice on this and would be happy to contribute to a project in the scikit-learn environment aiming to address similar issues, Thank you, Best, -- Roman

2 2

KMeans with cosine similarity
by JAGANADH G June 2, 2016

June 2, 2016

Hi Team, I was trying to use cosine similarity with KMeans. The code which I used to achieve is available here. https://gist.github.com/jaganadhg/b3f6af86ad99bf6e9bb7be21e5baa1b5 Is it the right way to achieve the same ? I know that cosine is not directly supported in sklearn KMeans. But after skimming through the code I was thinking that this will work ;-) -- ********************************** JAGANADH G http://jaganadhg.in *ILUGCBE* http://ilugcbe.org.in

2 1

Fwd: ValueError
by Ruchika Nayyar June 2, 2016

June 2, 2016

Thanks, Ruchika ---------------------------------------- Dr Ruchika Nayyar, Post Doctoral Fellow for ATLAS Collaboration University of Arizona Arizona, USA. -------------------------------------------- ---------- Forwarded message ---------- From: <scikit-learn-owner(a)python.org> Date: Wed, Jun 1, 2016 at 10:23 AM Subject: ValueError To: ruchika.work(a)gmail.com This list allows posts by subscribers only. Please subscribe at https://mail.python.org/mailman/listinfo/scikit-learn to post to the list. ---------- Forwarded message ---------- From: Ruchika Nayyar <ruchika.work(a)gmail.com> To: scikit-learn(a)python.org Cc: Date: Wed, 1 Jun 2016 10:23:14 -0700 Subject: ValueError Hi I am new to scikit-learn and while writing a python script to do a simple BDT using scikit-learn. I see error when I do this: from sklearn import datasets Traceback (most recent call last): File "bdt.py", line 12, in <module> from sklearn import datasets File "/Library/Python/2.7/site-packages/sklearn/__init__.py", line 57, in <module> from .base import clone File "/Library/Python/2.7/site-packages/sklearn/base.py", line 11, in <module> from .utils.fixes import signature File "/Library/Python/2.7/site-packages/sklearn/utils/__init__.py", line 10, in <module> from .murmurhash import murmurhash3_32 File "numpy.pxd", line 155, in init sklearn.utils.murmurhash (sklearn/utils/murmurhash.c:5029) ValueError: numpy.dtype has the wrong size, try recompiling I have already tried to uninstall numpy/scipy and pandas. They all are the latest and compatible version but something is not right. Can you tell me what I am doing wrong? Thanks, Ruchika ---------------------------------------- Dr Ruchika Nayyar, Post Doctoral Fellow for ATLAS Collaboration University of Arizona Arizona, USA. --------------------------------------------

9 21

Artificial neural network not learning lower values of the training sample
by muhammad waseem June 1, 2016

June 1, 2016

Hi All, I am trying to train an ANN but until now it is not learning the lower values of the training sample. I have tried using different python libraries to train ANN. The aim is to predict solar radiation from other weather parameters (regression problem). I think the ANN is confusing lower values (winter/cloudy days) with the night-time values (probably). I have tried the following but none of them worked; 1. Scaling data between different values e.g. [0,1],[-1,1] 2. Standardising data to have zero mean and unit variance 3. Shuffling the data 4. Increasing the training samples (from 3 years to 10 years) 5. Using different train functions 6. Trying different transfer functions 7. Using few input variables 8. Varying hidden layers and hidden layers' neurons Any idea what could be wrong or any directions to try? Thanks Kindest Regards Waseem

5 16