This question does not have a single "right" answer.
I'm interested in running Map Reduce algorithms, on a cluster, on Terabytes of data.
I want to learn more about the running time of said algorithms.
What books should I read?
I'm not interested in setting up Map Reduce clusters, or running standard algorithms. I want rigorous theoretical treatments or running time.
EDIT: The issue is not that map reduce changes running time. The issue is -- most algorithms do not distribute well to map reduce frameworks. I'm interested in algorithms that run on the map reduce framework.
Technically, there's no real different in the runtime analysis of MapReduce in comparison to "standard" algorithms - MapReduce is still an algorithm just like any other (or specifically, a class of algorithms that occur in multiple steps, with a certain interaction between those steps).
The runtime of a MapReduce job is still going to scale how normal algorithmic analysis would predict, when you factor in division of tasks across multiple machines and then find the maximum individual machine time required for each step.
That is, if you have a task which requires M map operations, and R reduce operations, running on N machines, and you expect that the average map operation will take m time and the average reduce operation r time, then you'll have an expected runtime of ceil(M/N)*m + ceil(R/N)*r time to complete all of the tasks in question.
Prediction of the values for M,R,m, and r are all something that can be accomplished with normal analysis of whatever algorithm you're plugging into MapReduce.
There are only two books that i know of that are published, but there are more in the works:
Pro hadoop and Hadoop: The Definitive Guide
Of these, Pro Hadoop is more of a beginners book, whilst The Definitive Guide is for those that know what Hadoop actually is.
I own The Definitive Guide and think its an excellent book. It provides good technical details on how the HDFS works, as well as covering a range of related topics such as MapReduce, Pig, Hive, HBase etc. It should also be noted that this book was written by Tom White who has been involved with the development of Hadoop for a good while, and now works at cloudera.
As far as the analysis of algorithms goes on Hadoop you could take a look at the TeraByte sort benchmarks. Yahoo have done a write up of how Hadoop performs for this particular benchmark: TeraByte Sort on Apache Hadoop. This paper was written in 2008.
More details about the 2009 results can be found here.
There is a great book about Data Mining algorithms applied to the MapReduce model.
It was written by two Stanford Professors and it if available for free:
http://infolab.stanford.edu/~ullman/mmds.html
Related
With respect specifically to CatBoost:
Under what scenarios might one want to use fewer than the max number of threads of one's CPU? I cannot find an answer to this.
Is there a fixed cost/overhead associated with each core utilized? I.e., is more always better for all data set types/sizes?
Do the answers to the questions above generalize to all machine learning algorithms?
I think that most of the reasons for changing the thread_count are not catboost specific. Other libraries like sklearn offer the same feature. Reasons for not running with all CPUs are:
Debugging: If there is a problem it might be handy to only have one thread thus making the process more simple.
You want other processes on your machine to have CPU power. Especially if you have a server for in-memory data analysis shared by a team of data scientists. Your colleagues won't be happy if you take all resources.
Your job is so small that it simply does not need all the resources.
Your parallelize in another way: For example you try different hyper parameters using cross validation. Then it would make sense to dedicate one CPU to training one model rather than training a model with with all CPUs and then move on to train the next model with all CPUs
I hope this answers question 1. This generalizes to other in-memory ml libraries like sklearn.
Regarding question 2 I'm not sure. CatBoost does the parallelisation somewhere in its C++ Code and uses it via Cython in the Python package. I assume it introduces some overhead (since distributed computing always introduces overhead) but it's probably not too much. You could find out by timing some experiments.
Following the Spark MLlib Guide we can read that Spark has two machine learning libraries:
spark.mllib, built on top of RDDs.
spark.ml, built on top of Dataframes.
According to this and this question on StackOverflow, Dataframes are better (and newer) than RDDs and should be used whenever possible.
The problem is that I want to use common machine learning algorithms (e.g: Frequent Pattern Mining,Naive Bayes, etc.) and spark.ml (for dataframes) don't provide such methods, only spark.mllib(for RDDs) provides this algorithms.
If Dataframes are better than RDDs and the referred guide recommends the use of spark.ml, why aren't common machine learning methods implemented in that lib?
What's the missing point here?
Spark 2.0.0
Currently Spark moves strongly towards DataFrame API with ongoing deprecation of RDD API. While number of native "ML" algorithms is growing the main points highlighted below are still valid and internally many stages are implemented directly using RDDs.
See also: Switch RDD-based MLlib APIs to maintenance mode in Spark 2.0
Spark < 2.0.0
I guess that the main missing point is that spark.ml algorithms in general don't operate on DataFrames. So in practice it is more a matter of having a ml wrapper than anything else. Even native ML implementation (like ml.recommendation.ALS use RDDs internally).
Why not implement everything from scratch on top of DataFrames? Most likely because only a very small subset of machine learning algorithms can actually benefit from the optimizations which are currently implemented in Catalyst not to mention be efficiently and naturally implemented using DataFrame API / SQL.
Majority of the ML algorithms require efficient linear algebra library not a tabular processing. Using cost based optimizer for linear algebra could be an interesting addition (I think that flink already has one) but it looks like for now there is nothing to gain here.
DataFrames API gives you very little control over the data. You cannot use partitioner*, you cannot access multiple records at the time (I mean a whole partition for example), you're limited to a relatively small set of types and operations, you cannot use mutable data structures and so on.
Catalyst applies local optimizations. If you pass a SQL query / DSL expression it can analyze it, reorder, apply early projections. All of that is that great but typical scalable algorithms require iterative processing. So what you really want to optimize is a whole workflow and DataFrames alone are not faster than plain RDDs and depending on an operation can be actually slower.
Iterative processing in Spark, especially with joins, requires a fine graded control over the number of partitions, otherwise weird things happen. DataFrames give you no control over partitioning. Also, DataFrame / Dataset don't provide native checkpoint capabilities (fixed in Spark 2.1) which makes iterative processing almost impossible without ugly hacks
Ignoring low level implementation details some groups of algorithms, like FPM, don't fit very well into a model defined by ML pipelines.
Many optimizations are limited to native types, not UDT extensions like VectorUDT.
There is one more problem with DataFrames, which is not really related to machine learning. When you decide to use a DataFrame in your code you give away almost all benefits of static typing and type inference. It is highly subjective if you consider it to be a problem or not but one thing for sure, it doesn't feel natural in Scala world.
Regarding better, newer and faster I would take a look at Deep Dive into Spark SQL’s Catalyst Optimizer, in particular the part related to quasiquotes:
The following figure shows that quasiquotes let us generate code with performance similar to hand-tuned programs.
* This has been changed in Spark 1.6 but it is still limited to default HashPartitioning
I am going to do a research project which involves predicting imminent failure of an engine using time data obtained from sensors. The data basically contains the readings of various embedded sensors every 10 minutes for many months. Such data is available for about 100 or so different units (all are the same engine model), along with the time of failure.
While I do have a reasonably good understanding of Machine Learning, I am at a loss of approaching this. I have done a few projects that involved static datasets (using SVMs, Neural Nets, Logistic Regression etc.) and even one on predicting time series. But this is quite different. While the project involves time data, it is hardly a matter of predicting the future values. Rather it is a case of anomaly detection on sequential time data.
Please could you give some ideas as to how I could approach it?
I'm particularly interested in Neural Networks/ Deep Learning, so any ideas on using them for this task would also be welcome. I would prefer to use Python or R, although I would be open to using something else if it was particularly geared for this sort of task.
Also could you give me some formal terms using which I could search for relevant literature?
Thanks
As a general comment, try hard to express everything that you know about the physical system in a model, then use that model for inference. I worked on such problems in my dissertation: Unified Prediction and Diagnosis in Engineering Systems by means of Distributed Belief Networks (see chapter 6). I can say more if you provide additional details about your problem domain.
Don't expect general machine learning models (neural networks, SVM, etc) to figure out the structure of the problem for you. Having the right form of the model is much, much more important than having a general model + lots of data -- this is the summary of my experience.
I've heard of a max flow min cut method for sharding or segmenting a graph database. Does someone have a sample cypher query that can do that say against the movielens dataset? Basically I want to segment users into different shards/clusters based on what they like so maybe the min cuts can naturally find clusters of users around the genres say Horror, Drama, or maybe it will create non-intuitive clusters/segments like hipster/romantics and conservative/comedy/horror groups.
my short answer is no, sorry I don't know how you would express that.
my longer answer is even if this were possible - which it very well may be - I would advise against it.
multiple algorithms 'do' min-cut max-flow, these will all have different performance characteristics and, because clustering is computationally expensive, I'd guess you want control over the specific algorithm implementation used.
Cypher is a declarative language, you specify what you're looking for but not how to do it, and it will be difficult to specify such a complex problem in a way that the Cypher engine can figure out what you're trying to do. that will make it hard for Cypher (or any declarative language engine) to produce an efficient query plan.
my suggestion is find the specific algorithm you wish to use and implement it using the Neo4j Java API.
if you're running Neo4j in embedded mode you're done at that point. if you're running Neo4j server you'll then just have to run that code as an Unmanaged Server Extension
AFAIK you're after 'Community Detection' algorithms. There are non-overlapping (communities do not overlap) and overlapping variants, where non-overlapping is generally easier to implement and understand. Common algorithms are:
Non-overlapping: Louvain
Overlapping: Label Propagation Algorithm (LPA) (typically non-overlapping, but there are extensions to make it overlapping)
Here are a few C++ code examples for the algorithms: Louvain, Oslom (overlapping), LPA (non-overlapping), and Infomap)
And if you want bleeding edge I was recommended the SCD algorithm
Academic paper: "High Quality, Scalable and Parallel Community Detection for Large Real Graphs"
C++ implementation
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 2 years ago.
Improve this question
In the beginning, I would like to describe my current position and the goal that I would like to achieve.
I am a researcher dealing with machine learning. So far have gone through several theoretical courses covering machine learning algorithms and social network analysis and therefore have gained some theoretical concepts useful for implementing machine learning algorithms and feed in the real data.
On simple examples, the algorithms work well and the running time is acceptable whereas the big data represent a problem if trying to run algorithms on my PC. Regarding the software I have enough experience to implement whatever algorithm from articles or design my own using whatever language or IDE (so far have used Matlab, Java with Eclipse, .NET...) but so far haven't got much experience with setting-up infrastructure. I have started to learn about Hadoop, NoSQL databases, etc, but I am not sure what strategy would be the best taking into consideration the learning time constraints.
The final goal is to be able to set-up a working platform for analyzing big data with focusing on implementing my own machine learning algorithms and put all together into production, ready for solving useful question by processing big data.
As the main focus is on implementing machine learning algorithms I would like to ask whether there is any existing running platform, offering enough CPU resources to feed in large data, upload own algorithms and simply process the data without thinking about distributed processing.
Nevertheless, such a platform exists or not, I would like to gain a picture big enough to be able to work in a team that could put into production the whole system tailored upon the specific customer demands. For example, a retailer would like to analyze daily purchases so all the daily records have to be uploaded to some infrastructure, capable enough to process the data by using custom machine learning algorithms.
To put all the above into simple question: How to design a custom data mining solution for real-life problems with main focus on machine learning algorithms and put it into production, if possible, by using the existing infrastructure and if not, design distributed system (by using Hadoop or whatever framework).
I would be very thankful for any advice or suggestions about books or other helpful resources.
First of all, your question needs to define more clearly what you intend by Big Data.
Indeed, Big Data is a buzzword that may refer to various size of problems. I tend to define Big Data as the category of problems where the Data size or the Computation time is big enough for "the hardware abstractions to become broken", which means that a single commodity machine cannot perform the computations without intensive care of computations and memory.
The scale threshold beyond which data become Big Data is therefore unclear and is sensitive to your implementation. Is your algorithm bounded by Hard-Drive bandwidth ? Does it have to feet into memory ? Did you try to avoid unnecessary quadratic costs ? Did you make any effort to improve cache efficiency, etc.
From several years of experience in running medium large-scale machine learning challenge (on up to 250 hundreds commodity machine), I strongly believe that many problems that seem to require distributed infrastructure can actually be run on a single commodity machine if the problem is expressed correctly. For example, you are mentioning large scale data for retailers. I have been working on this exact subject for several years, and I often managed to make all the computations run on a single machine, provided a bit of optimisation. My company has been working on simple custom data format that allows one year of all the data from a very large retailer to be stored within 50GB, which means a single commodity hard-drive could hold 20 years of history. You can have a look for example at : https://github.com/Lokad/lokad-receiptstream
From my experience, it is worth spending time in trying to optimize algorithm and memory so that you could avoid to resort to distributed architecture. Indeed, distributed architectures come with a triple cost. First of all, the strong knowledge requirements. Secondly, it comes with a large complexity overhead in the code. Finally, distributed architectures come with a significant latency overhead (with the exception of local multi-threaded distribution).
From a practitioner point of view, being able to perform a given data mining or machine learning algorithm in 30 seconds is one the key factor to efficiency. I have noticed than when some computations, whether sequential or distributed, take 10 minutes, my focus and efficiency tend to drop quickly as it becomes much more complicated to iterate quickly and quickly test new ideas. The latency overhead introduced by many of the distributed frameworks is such that you will inevitably be in this low-efficiency scenario.
If the scale of the problem is such that even with strong effort you cannot perform it on a single machine, then I strongly suggest to resort to on-shelf distributed frameworks instead of building your own. One of the most well known framework is the MapReduce abstraction, available through Apache Hadoop. Hadoop can be run on 10 thousands nodes cluster, probably much more than you will ever need. If you do not own the hardware, you can "rent" the use of a Hadoop cluster, for example through Amazon MapReduce.
Unfortunately, the MapReduce abstraction is not suited to all Machine Learning computations.
As far as Machine Learning is concerned, MapReduce is a rigid framework and numerous cases have proved to be difficult or inefficient to adapt to this framework:
– The MapReduce framework is in itself related to functional programming. The
Map procedure is applied to each data chunk independently. Therefore, the
MapReduce framework is not suited to algorithms where the application of the
Map procedure to some data chunks need the results of the same procedure to
other data chunks as a prerequisite. In other words, the MapReduce framework
is not suited when the computations between the different pieces of data are
not independent and impose a specific chronology.
– MapReduce is designed to provide a single execution of the map and of the
reduce steps and does not directly provide iterative calls. It is therefore not
directly suited for the numerous machine-learning problems implying iterative
processing (Expectation-Maximisation (EM), Belief Propagation, etc.). The
implementation of these algorithms in a MapReduce framework means the
user has to engineer a solution that organizes results retrieval and scheduling
of the multiple iterations so that each map iteration is launched after the reduce
phase of the previous iteration is completed and so each map iteration is fed
with results provided by the reduce phase of the previous iteration.
– Most MapReduce implementations have been designed to address production needs and
robustness. As a result, the primary concern of the framework is to handle
hardware failures and to guarantee the computation results. The MapReduce efficiency
is therefore partly lowered by these reliability constraints. For example, the
serialization on hard-disks of computation results turns out to be rather costly
in some cases.
– MapReduce is not suited to asynchronous algorithms.
The questioning of the MapReduce framework has led to richer distributed frameworks where more control and freedom are left to the framework user, at the price of more complexity for this user. Among these frameworks, GraphLab and Dryad (both based on Direct Acyclic Graphs of computations) are well-known.
As a consequence, there is no "One size fits all" framework, such as there is no "One size fits all" data storage solution.
To start with Hadoop, you can have a look at the book Hadoop: The Definitive Guide by Tom White
If you are interested in how large-scale frameworks fit into Machine Learning requirements, you may be interested by the second chapter (in English) of my PhD, available here: http://tel.archives-ouvertes.fr/docs/00/74/47/68/ANNEX/texfiles/PhD%20Main/PhD.pdf
If you provide more insight about the specific challenge you want to deal with (type of algorithm, size of the data, time and money constraints, etc.), we probably could provide you a more specific answer.
edit : another reference that could prove to be of interest : Scaling-up Machine Learning
I had to implement a couple of Data Mining algorithms to work with BigData too, and I ended up using Hadoop.
I don't know if you are familiar to Mahout (http://mahout.apache.org/), which already has several algorithms ready to use with Hadoop.
Nevertheless, if you want to implement your own Algorithm, you can still adapt it to Hadoop's MapReduce paradigm and get good results. This is an excellent book on how to adapt Artificial Intelligence algorithms to MapReduce:
Mining of Massive Datasets - http://infolab.stanford.edu/~ullman/mmds.html
This seems to be an old question. However given your usecase, the main frameworks focusing on Machine Learning in Big Data domain are Mahout, Spark (MLlib), H2O etc. However to run Machine Learning algorithms on Big Data you have to convert them to parallel programs based on Map Reduce paradigm. This is a nice article giving a brief introduction to major (not all) big Data frameworks:
http://www.codophile.com/big-data-frameworks-every-programmer-should-know/
I hope this will help.