Related
I connect Tableau to R and execute an R function for recommending products. When R ends, the return value is a string which will have all products details, like below:
ID|Existing_Prod|Recommended_Prod\nC001|NA|PROD008\nC002|PROD003|NA\nF003|NA|PROD_ABC\nF004|NA|PROD_ABC1\nC005|PROD_ABC2|NA\nC005|PRODABC3|PRODABC4
(Each line separated by \n indicating end of line)
On Tableau, I display the calculated field which is as below:
ID|Existing_Prod|Recommended_Prod
C001|NA|PROD008
C002|PROD003|NA
F003|NA|PROD_ABC
F004|NA|PROD_ABC1
C005|PROD_ABC2|NA
C005|PRODABC3|PRODABC4
Above data reaches Tableau through a calculated field as a single string which I want to split based on pipeline ('|'). Now, I need to split this into three columns, separated by the pipeline.
I used Split function on the calculated field :
SPLIT([R_Calculated_Field],'|',1)
SPLIT([R_Calculated_Field],'|',2)
SPLIT([R_Calculated_Field],'|',3)
But the error says "SPLIT function cannot be applied on Table calculations", which is self explanatory. Are there any alternatives to solve this ?? I googled to check for best practices to handle integration between R and Tableau and all I could find was simple kmeans clustering codes.
Make sure you understand how partitioning and addressing work for table calcs. Table calcs pass vectors of arguments to the R script, and receive a single vector in response. The cardinality of those vectors depends on the partitioning of the table calc. You can view that by editing the table calc, clicking specific dimensions. The fields that are not checked determine the partitioning - and thus the cardinality of the arguments you send and receive from R
This means it might be tricky to map your problem onto this infrastructure. Not necessarily impossible. It was designed to send a series of vector arguments with one cell per partitioning dimension, say, Manufacturer and get back one vector with one result per Manufacturer (or whatever combination of fields partition your data for the table calc). Sounds like you are expecting an arbitrary length list of recommendations. It shouldn’t be too hard to have your R script turn the string into a vector before returning, but the size of the vector has to make sense.
As an example of an approach that fits this model more easily, say you had a Tableau view that had one row per Product (and you had N products) - and some other aggregated measure fields in the view per Product. (In Tableau speak, the view’s level of detail is at the Product level.)
It would be straightforward to pass those measures as a series of argument vectors to R - each vector having N values, and then have R return a vector of reals of length N where the value returned at each location was a recommender score for the product at that position. (Which is why the ordering aka addressing of the vectors also matters)
Then you could filter out low scoring products from the view and visually distinguish highly recommended products.
So the first step to understanding R integration is to understand how table calcs operate with partitioning and addressing and to think in terms of vectors of fixed lengths passed in both directions.
If this model doesn’t support your use case well, you might be able to do something useful with URL actions or the JavaScript API.
Trying to create an array from an xyz data file. The data file is arranged so that x,y,z of each atom is on a new line and I want the array to reflect this.
Then to use this array to find find the distance from each atom in the list with all the others.
To do this the array has been copied such that atom1 & atom2 should be identical to the input file.
length is simply the number of atoms in the list.
The write statement: WRITE(20,'(3F12.9)') atom1 actually gives the matrix wanted but when I try to find individual elements they're all wrong!
Any help would be really appreciated!
Thanks guys.
DOUBLE PRECISION, DIMENSION(:,:), ALLOCATABLE ::atom1,atom2'
ALLOCATE(atom1(length,3),atom2(length,3))
READ(10,*) ((atom1(i,j), i=1,length), j=1,3)
atom2=atom1
distn=0
distc=0
DO n=1,length
x1=atom1(n,1)
y1=atom1(n,2) !1st atom
z1=atom1(n,3)
DO m=1,length
x2=atom2(m,1)
y2=atom2(m,2) !2nd atom
z2=atom2(m,3)`
Your READ statement reads all the x coordinates for all atoms from however many records, then all the y coordinates, then all the z coordinates. That's inconsistent with your description of the input file. You have the nesting of the io-implied-do's in the READ statement around the wrong way - it should be ((atom1(i,j),j=1,3),i=1,length).
Similarly, as per the comment, your diagnostic write mislead you - you were outputting all x ordinates, followed by all y ordinates, etc. Array element order of a whole array reference varies the first (leftmost) dimension fastest (colloquially known as column major order).
(There are various pitfalls associated with list directed formatting that mean I wouldn't recommend it for production code (or perhaps for input specifically written with the knowledge of and defence against those pitfalls). One of those pitfalls is that the READ under list directed formatting will pull in as many records as it requires to satisfy the input list. You may have detected the problem earlier if you were using an explicit format that nominated the number of fields per record.)
I want to test some of the newer sparse linear solvers and I want to know if there is a fast way of filling in the matrix. The format I'm interested is CSR (http://goo.gl/hLXYd). Let's say the matrix, in CSR format, is given by:
values(num non-zero elements)
columns(num non-zero elements)
rowIndex(num rows + 1)
The sparse matrix under consideration derives from networks. So, I have thousands of nodes and some of them are connected between them by lines. So, the matrix is structurally symmetric. Each connection (i,j) adds something to the diagonal terms (i,i) and (j,j) and to the off-diagonal (i,j) and (j,i). I could have several connections between the same nodes (i,j,1), (i,j,2)... So, I might need to revisit the 2 diagonal and 2 off-diagonal elements more than once.
I know I can get the beginning of the row by doing rowIndex(i). Then, I would have to run through the elements columns(rowIndex(i):rowIndex(i+1)-1) to find where is j situated.
The question:
Is there a way of accessing the elements faster, while in CSR format, without having to do a search every time I want to update an element?
Some clarifications:
I just need to fill in the matrix from scratch. The matrix is structurally symmetric and not really symmetric. The values saved have to do with network data (impedances, resistances etc), they have real values. In general Value(i,j)<>Value(j,i). I have tuples of the form (name1,i1,j1,value1), (name2,i2,j2,value2) etc. These tuples are not sorted, and 2 tuples can refer to the same i,j values, meaning they need to be added
Thanks in advance!
What you have is so called triplet sparse format. Creation of CRS, including removing duplicate entries and summing the values, can be implemented very efficiently. Before programing it yourself, have a look at the SuiteSparse library. It is written in C, but I'm sure you will understand the principle. What interests you is the cholmod_triplet.c file, which implements the functionality you need.
Essentially, the conversion is performed using two phase bucket sort on your row and column indices. This algorithm has linear complexity, which is important if you are interested in processing large data sets.
Edit If you want to skip explicit creation of the triplet format all together, you can do that by generating the (row, col) connectivities on the fly and adding them to a dynamic sparse structure. I usually do it using insertion sort and sorted lists, which is in practice the fastest. It is also faster than triplet to CRS conversion, and uses much less memory. The method goes as follows:
if you know approximately, how many non-zero entries there are in every row, for every row you pre-allocate an array of (empty) column indices, and a separate array for the values (not linked list, but a simple array) of that size. Something like
static_lists_cols[row] = malloc(sizeof(int)*expected_number_of_non_zeros)
static_lists_vals[row] = malloc(sizeof(double)*expected_number_of_non_zeros)
If you do not know that, you choose an initial size and reallocate as needed (using some block size large enough to avoid reallocation overhead) when the row lists are full.
for every (row, col) pair you insert the col into the sorted list corresponding to row using insertion sort. For small (up to a few hundred) non-zeros per row linear search is the fastest. For larger number of non-zeros per row you can use bisection to locate the correct place to insert the col index.
col is inserted into rowth sorted list by moving the non-zero entries with higher column index in the sorted list. This is cache-friendly, since the rows are in practice small enough to fit into any cache nowadays.
After you finish you need to assemble the individual sorted lists into a valid CRS structure by copying the individual row lists into the final columns. The same with values.
You could actually avoid the last step by pre-allocating a static 'array of lists' if you are ok that some of the rows can have zero entries. You will hence have a constant number of entries per row, some of which might be zero. Sometimes that is ok.
This method is faster than using triplet to sparse conversion, at least for FEM models, for which I use it. The general reason is that memory bandwidth is the bottleneck here, and the above scheme uses much less memory:
creating the triplet format takes time, and you need to write the triplets to memory
conversion to CRS requires reading and writing the triplets at least once to sort them (actually a bit more than once, if you look at the algorithm. You sort twice, and you need auxiliary data structures.)
depending on the connectivity structure, you may end up having a large number of (row, col) duplicates in the triplet format, which are removed during the assembly by adding the corresponding values. This overhead does not exist in the method above - if the col already exists in the row list, you simply update the corresponding value.
updating the sorted lists can be done in parallel if you assign row ranges to individual workers. No communication, nor synchronization is needed. Assuring load balancing is another story...
Have a look at a performance comparison of using those two methods (Figure 1) for triangular elements in 2D. Note that the performance difference depends on the ratio of the number of entries in the triplet to assembled sparse matrix format (Table 2). But in general, the method is never worse than triplet to crs conversion, and triplets need to be created in the first place. You can also download a MATLAB MEX function sparse_create, which is a part of mutils package (see the downloads section).
Your question seems to confuse 2 rather different questions:
What is a fast way of creating a matrix in CSR form ?
Is there a faster way of reading values from a matrix already stored in CSR form ? (Faster, that is, than the straightforward approach you describe)
So here are 2 answers:
In general, read the network data from whatever form it is in into something like a dictionary of keys (other intermediate forms are available and may be more appealing to you for speed or other reasons); then turn that intermediate structure into the CSR form of the matrix. More on this below.
I don't believe so, not with a matrix stored in CSR form. This relative slowness of access is part of the price you pay for saving space. You've traded time for space, or space for time, depending on your point of view.
Your description of your input data suggests that you should consider devising your own intermediate form into which to marshal the raw data. Since your adjacency matrix is symmetric you only need to store, in any form, half of it. Further, you probably don't need to store the elements along the main diagonal -- I'm guessing either that node i is always connected to node i or never so that the nature of the network determines the value stored at (i,i). I'm a little uncertain of the information you want to store at each node of the matrix, is it the number of connections between i and j or something else ?
I have about 42,000 lists of 24 random numbers, all in the range [0, 255]. For example, the first list might be [32, 15, 26, 27, ... 11]. The second list might be [44, 44, 18, 19, .. 113]. How can I choose a number from each of the lists so that (so I will end up with a new list of about 42,000 numbers) such that this new list is most compressible using ZIP?
-- this question has to do with math, data compression
The ZIP file format uses DEFLATE for its compression algorithm. So you need to consider how that algorithm works and pick data such that the algorithm finds it easy to compress. According to the wikipedia article, there are two stages of compression. The first uses LZ77 to find repeated sections of data and replace them with short references. The second uses Huffman coding to take the remaining data and strip out redundancy across the whole block. This is called entropic coding - if the information isn't very random (has low entropy) the code replaces common things with short symbols, increasing the entropy.
In general, then, lists with lots of repeated runs (i.e., [111,2,44,93,111,2,44,93...]) will compress well in the first pass. Lists with lots of repeated numbers within other random stuff (i.e., [111,34,43,50,111,34,111,111,2,34,22,60,111,98,2], where 34 and 111 show up often) will compress well in the second pass.
To find suitable numbers, I think the easiest thing to do is just sort each list, then merge them, keeping the merge sorted, until you get to 42000 output numbers. You'll get runs as they happen. This won't be optimal, you might have the number 255 in each input list and you'd miss them using this technique, but it would be easy.
Another approach would be to histogram the numbers into 256 bins. Any bins that stand out indicate numbers that should be grouped. After that, I guess you have to search for sequences. Again, sorting the inputs will probably make this easier.
I just noticed you had the constraint that you have to pick one number from each list. So in both cases you could sort each list then remove duplicates.
Additionally, Huffman codes can be generated using a tree, so I wonder if there's some magic tree structure you could put the numbers into that would automatically give the right answer.
This smells NP-complete to me, but I am nowhere near able to prove it. On the outside, there are approximately 7.45e+57968 (!) possible configurations to test. It doesn't seem that you can opt out of a particular configuration early, as an incompressible initial section could be greatly compressible later on.
My best guess for "good" compression would be to count the number of occurrences of each number across the entire million-element set and select from each list the numbers with the most occurrences. For example, if every list has 42 present in it, selecting that only would give you a very-compressible array of 42,000 instances of the same value.
I need to automatically match product names (cameras, laptops, tv-s etc) that come from different sources to a canonical name in the database.
For example "Canon PowerShot a20IS", "NEW powershot A20 IS from Canon" and "Digital Camera Canon PS A20IS"
should all match "Canon PowerShot A20 IS". I've worked with levenshtein distance with some added heuristics (removing obvious common words, assigning higher cost to number changes etc), which works to some extent, but not well enough unfortunately.
The main problem is that even single-letter changes in relevant keywords can make a huge difference, but it's not easy to detect which are the relevant keywords. Consider for example three product names:
Lenovo T400
Lenovo R400
New Lenovo T-400, Core 2 Duo
The first two are ridiculously similar strings by any standard (ok, soundex might help to disinguish the T and R in this case, but the names might as well be 400T and 400R), the first and the third are quite far from each other as strings, but are the same product.
Obviously, the matching algorithm cannot be a 100% precise, my goal is to automatically match around 80% of the names with a high confidence.
Any ideas or references is much appreciated
I think this will boil down to distinguishing key words such as Lenovo from chaff such as New.
I would run some analysis over the database of names to identify key words. You could use code similar to that used to generate a word cloud.
Then I would hand-edit the list to remove anything obviously chaff, like maybe New is actually common but not key.
Then you will have a list of key words that can be used to help identify similarities. You would associate the "raw" name with its keywords, and use those keywords when comparing two or more raw names for similarities (literally, percentage of shared keywords).
Not a perfect solution by any stretch, but I don't think you are expecting one?
The key understanding here is that you do have a proper distance metric. That is in fact not your problem at all. Your problem is in classification.
Let me give you an example. Say you have 20 entries for the Foo X1 and 20 for the Foo Y1. You can safely assume they are two groups. On the other hand, if you have 39 entries for the Bar X1 and 1 for the Bar Y1, you should treat them as a single group.
Now, the distance X1 <-> Y1 is the same in both examples, so why is there a difference in the classification? That is because Bar Y1 is an outlier, whereas Foo Y1 isn't.
The funny part is that you do not actually need to do a whole lot of work to determine these groups up front. You simply do an recursive classification. You start out with node per group, and then add the a supernode for the two closest nodes. In the supernode, store the best assumption, the size of its subtree and the variation in it. As many of your strings will be identical, you'll soon get large subtrees with identical entries. Recursion ends with the supernode containing at the root of the tree.
Now map the canonical names against this tree. You'll quickly see that each will match an entire subtree. Now, use the distances between these trees to pick the distance cutoff for that entry. If you have both Foo X1 and Foo Y1 products in the database, the cut-off distance will need to be lower to reflect that.
edg's answer is in the right direction, I think - you need to distinguish key words from fluff.
Context matters. To take your example, Core 2 Duo is fluff when looking at two instances of a T400, but not when looking at a a CPU OEM package.
If you can mark in your database which parts of the canonical form of a product name are more important and must appear in one form or another to identify a product, you should do that. Maybe through the use of some sort of semantic markup? Can you afford to have a human mark up the database?
You can try to define equivalency classes for things like "T-400", "T400", "T 400" etc. Maybe a set of rules that say "numbers bind more strongly than letters attached to those numbers."
Breaking down into cases based on manufacturer, model number, etc. might be a good approach. I would recommend that you look at techniques for term spotting to try and accomplish that: http://www.worldcat.org/isbn/9780262100854
Designing everything in a flexible framework that's mostly rule driven, where the rules can be modified based on your needs and emerging bad patterns (read: things that break your algorithm) would be a good idea, as well. This way you'd be able to improve the system's performance based on real world data.
You might be able to make use of a trigram search for this. I must admit I've never seen the algorithm to implement an index, but have seen it working in pharmaceutical applications, where it copes very well indeed with badly misspelt drug names. You might be able to apply the same kind of logic to this problem.
This is a problem of record linkage. The dedupe python library provides a complete implementation, but even if you don't use python, the documentation has a good overview of how to approach this problem.
Briefly, within the standard paradigm, this task is broken into three stages
Compare the fields, in this case just the name. You can use one or more comparator for this, for example an edit distance like the Levenshtein distance or something like the cosine distance that compares the number of common words.
Turn an array fo distance scores into a probability that a pair of records are truly about the same thing
Cluster those pairwise probability scores into groups of records that likely all refer to the same thing.
You might want to create logic that ignores the letter/number combination of model numbers (since they're nigh always extremely similar).
Not having any experience with this type of problem, but I think a very naive implementation would be to tokenize the search term, and search for matches that happen to contain any of the tokens.
"Canon PowerShot A20 IS", for example, tokenizes into:
Canon
Powershot
A20
IS
which would match each of the other items you want to show up in the results. Of course, this strategy will likely produce a whole lot of false matches as well.
Another strategy would be to store "keywords" with each item, such as "camera", "canon", "digital camera", and searching based on items that have matching keywords. In addition, if you stored other attributes such as Maker, Brand, etc., you could search on each of these.
Spell checking algorithms come to mind.
Although I could not find a good sample implementation, I believe you can modify a basic spell checking algorithm to comes up with satisfactory results. i.e. working with words as a unit instead of a character.
The bits and pieces left in my memory:
Strip out all common words (a, an, the, new). What is "common" depends on context.
Take the first letter of each word and its length and make that an word key.
When a suspect word comes up, looks for words with the same or similar word key.
It might not solve your problems directly... but you say you were looking for ideas, right?
:-)
That is exactly the problem I'm working on in my spare time. What I came up with is:
based on keywords narrow down the scope of search:
in this case you could have some hierarchy:
type --> company --> model
so that you'd match
"Digital Camera" for a type
"Canon" for company and there you'd be left with much narrower scope to search.
You could work this down even further by introducing product lines etc.
But the main point is, this probably has to be done iteratively.
We can use the Datadecision service for matching products.
It will allow you to automatically match your product data using statistical algorithms. This operation is done after defining a threshold score of confidence.
All data that cannot be automatically matched will have to be manually reviewed through a dedicated user interface.
The online service uses lookup tables to store synonyms as well as your manual matching history. This allows you to improve the data matching automation next time you import new data.
I worked on the exact same thing in the past. What I have done is using an NLP method; TF-IDF Vectorizer to assign weights to each word. For example in your case:
Canon PowerShot a20IS
Canon --> weight = 0.05 (not a very distinguishing word)
PowerShot --> weight = 0.37 (can be distinguishing)
a20IS --> weight = 0.96 (very distinguishing)
This will tell your model which words to care and which words to not. I had quite good matches thanks to TF-IDF.
But note this: a20IS cannot be recognized as a20 IS, you may consider to use some kind of regex to filter such cases.
After that, you can use a numeric calculation like cosine similarity.