Search Results

Search found 2242 results on 90 pages for 'fuzzy comparison'.

Page 11/90 | < Previous Page | 7 8 9 10 11 12 13 14 15 16 17 18  | Next Page >

• Is memcached a dinosaur in comparison to Redis?

  - by Industrial

  Hi everyone, I have worked quite a bit with memcached the last weeks and just found out about Redis. When I read this part of their readme, I suddenly got a warm, cozy feeling in my stomach: Redis can be used as a memcached on steroids because is as fast as memcached but with a number of features more. Like memcached, Redis also supports setting timeouts to keys so that this key will be automatically removed when a given amount of time passes. This sounds amazing. I'd also found this page with benchmarks: http://www.ruturaj.net/redis-memcached-tokyo-tyrant-mysql-comparison So, honestly - Is memcache really that old dinousaur that is a bad choice from a performance perspective when compared to this newcomer called Redis? I haven't heard lot about Redis previously, thereby the approach for my question!

  Read the article

• Why is Rails date comparison not working?

  - by revgum

  What am I missing here, it's driving me crazy.. >> user.current_login_at.utc > 24.hours.ago.utc => false >> 24.hours.ago.utc => Mon May 17 18:46:16 UTC 2010 >> user.current_login_at.utc => Mon May 17 15:47:44 UTC 2010 user.current_login_at was 27 hours ago, yet the greater than comparison says it was not greater than 24 hours ago. It leaves me scratching my head..

  Read the article

• Linq: comparison with nonexisting value should not cause the exception, but it does

  - by Seacat

  Hello, Given I don't know if the parameter will be null or not and I want to use it in the following way: if Param != null then compare with its id if Param == null then compare with null. Something like that: var c = from cat in context.Categories where ParamCat != null && cat.ParentId == ParamCat.Id || ParamCat == null && cat.ParentId == null select c; If ParamCat is null as soon as I try to get something from c (for example c.Count()) it throws the exception. Usually when we use some kind of condition it stop comparison as soon as condition fail especially if we use AND. For example this code will not cause any exception: if (ParamCat != null && cat.ParentId == RaramCat.Id) { } If so, why the linq code above throw exception? (Null reference) Thanks

  Read the article

• Fast comparison of char arrays?

  - by StackedCrooked

  I'm currently working in a codebase where IPv4 addresses are represented as pointers to u_int8. The equality operator is implemented like this: bool Ipv4Address::operator==(const u_int8 * inAddress) const { return (*(u_int32*) this->myBytes == *(u_int32*) inAddress); } This is probably the fasted solution, but it causes the GCC compiler warning: ipv4address.cpp:65: warning: dereferencing type-punned pointer will break strict-aliasing rules How can I rewrite the comparison correctly without breaking strict-aliasing rules and without losing performance points? I have considered using either memcmp or this macro: #define IS_EQUAL(a, b) \ (a[0] == b[0] && a[1] == b[1] && a[2] == b[2] && a[3] == b[3]) I'm thinking that the macro is the fastest solution. What do you recommend?

  Read the article

• Comparison tool with easy line insertion

  - by Miro Kropacek

  Back in good old days I used to use a tool for file comparison with one incredible feature -- you open file1, file2, see a difference, no magic here. But then you could insert an empty line(s) into file1 with one keyboard combo and into file2 with another keyboard combo. So you could easily adjust how are C / asm function aligned in case diff engine failed to recognize similar stuff. Of course, after the adjust (insertion / removal of one or more lines in either file) whole diff was "recalculated". I fail to find similar feature in diff, KDiff, ... I'd prefer Linux app but I'm OK with Windows app as last resort... Thanks for any hint!

  Read the article

• find all occurrences of comparison with == in visual studio

  - by Emiswelt

  Hi there I made the mistake of using == for comparing IP addresses instead of using the equals() method of the IPAddress class in C#, which will result in the comparison of references instead of values. Since the solution I am currently working on is very large for a one-man project ( 100.000 lines of source code), I am very sure that I still have some of these wrong statements in my code. Is there any possibility to tell Visual Studio to find all occurrences of == operations on a specific class for me, so that I can find and clean up the bugged comparisons? with best regards, emi

  Read the article

• Smarter "Confirm File Replace" dialog on Windows

  - by Niall C.

  In Windows (XP at least), when you drag a file to a folder that already has a file with the same name, you get a dialog like this: Does anyone know of a replacement for that dialog that adds a Compare button (preferably configurable to use the comparison tool of my choice)?

  Read the article

• Bash or Zsh - which one and why?

  - by Andrew

  So, the question pretty much says it all. I'm on Snow Leopard, and I do a lot of web development, particularly in Rails 3 which makes heavy use of the console. I've seen some notable bloggers etc. mention Zsh as their preference over Bash, but I don't know what difference it would make. Could anyone give me a good comparison of what difference there is and what might make one prefer one option or the other? Thanks!

  Read the article

• How to compare mp3, flac audio data in a file, ignoring header data (ID3 tag) etc.?

  - by Rob

  I've backed up some audio files up in 2 places and added ID3 tags into one backup but not the other, since time has passed my own memory has faded on whether the backups are actually the same, but now one has ID3 data and the other doesn't, basic binary compare will fail and inspection will be cumbersome. Is there a tool to compare just the audio data (not the header, ID3) in mp3s, flac files, and other files using header data such as ID3. started a thread on beyond compare here: http://www.scootersoftware.com/vbulletin/showthread.php?t=7413 would consider other comparison software that does this task

  Read the article

• compare windows server for patch/update/hotfix installs

  - by user12002221

  Are there any tools that can be used to connect to windows 2008 servers, and get a comparison of the installed patches/updates on the servers, showing what is installed on one and not on the other? This is to help isolate an issue we are seeing on a specific windows server, in a load balanced setup. There is a certain performance/locking issue, which is mitigated whenever one of the servers is disabled. Please share, if you have any suggestions. Thanks in advance!

  Read the article

• Inverse Logistic Function / Reverse Sigmoid Function

  - by Chanq

  I am currently coding up a fuzzy logic library in java. I have found the equations for all the standard functions - Grade, inverseGrade, Triangle, Trapezoid, Gaussian. However, I can't find the inverse of the sigmoid/ logistic function. The way I have written the logistic function is java is : //f(x) = 1/(1+e(-x)) public double logistic(double x){ return (1/(1+(Math.exp(-x))); } But I can't work out or find the inverse anywhere. My algebraic/calculus abilities are fairly limited, hence why I haven't been able to work out the inverse of the function. Any hints or pointers would be a big help. Thanks

  Read the article

• Looking for merge tool with very good in-line-comparison support

  - by peter p

  I have seen this topic discussed several times, but emphasis is on "very good in-line-comparison" here, which was not really covered by those threads. E.g. I would like the tool to recognize and highlight that the resource "colorpicker_newstring" has been added when comparing the following two blocks. WinMerge and Kdiff both fail... Does anyone know a software that does not? I am using Windows. Ah, and I'd prefer free/OS software, of course. Thanks a lot in advance, Peter File A: <resource name="colorpicker_title">Color picker</resource> <resource name="colorpicker_apply">Apply</resource> <resource name="colorpicker_transparent">Transparent</resource> <resource name="colorpicker_htmlcolor">HTML color</resource> <resource name="colorpicker_red">Red</resource> <resource name="colorpicker_newstring">New String</resource> <resource name="colorpicker_green">Green</resource> <resource name="colorpicker_blue">Blue</resource> <resource name="colorpicker_alpha">Alpha</resource> <resource name="colorpicker_recentcolors">Recently used colors</resource> File B: <resource name="colorpicker_title">Sélecteur de couleur</resource> <resource name="colorpicker_apply">Appliquer</resource> <resource name="colorpicker_transparent">Transparent</resource> <resource name="colorpicker_htmlcolor">Couleur HTML</resource> <resource name="colorpicker_red">Rouge</resource> <resource name="colorpicker_green">Vert</resource> <resource name="colorpicker_blue">Bleu</resource> <resource name="colorpicker_alpha">Alpha</resource> <resource name="colorpicker_recentcolors">Couleurs récentes</resource>

  Read the article

• convert htmlelement to string for comparison javascript

  - by Jamex

  Hi, I am using a function that obtains a target element id at onclick. Example, if I click on the text element that has the id of 'help'. var click = (e && e.target) || (event && event.srcElement); The var click would contain the ref to the id of "help". I want to compare the var click to the string 'help' using the if statement below. if (click == 'about') {do something} The comparison does not work because the var click is not a string. When I use the alert(click) to debug, it shows click as "object HTMLElement". How would you compare whether the id 'help' is obtained from var click? I could write out something like if (click == document.getElementById('help')) {do something} but that would make a long statement. also if the var click is document.getElementById('help'), how would you make a new var "show" as document.getElementById('showhelp') basically, I want to use the same function to generate dynamic responses to each element that was clicked on, and not having to create a separate function for each element.

  Read the article

• What free expert system can You recommend (with higher functionality then CLIPS)?

  - by Martin

  Hi, I'm trying to find best free expert system, with the highest functionality. I know about CLIPS, but is there another system, for example being able to accept percent of confidence for each rule (fuzzy logic). I need it to know will I be able to do fast a short project using expert system, with highest functionality. But anyways, it's interesting is there an open source program that aims to gather different AI methods (whitch there are plenty of), and use them together. So I would be extremely thankeful for any info about more robust CLIPS, or similar programs. Thanks!

  Read the article

• Draw a comparison between an integer in a specific row and a count

  - by XCoderX

  This is a follow-up question to this one: Check for specific integer in a row WHERE user = $name I want a user to be able to comment on my site for exactly five times a day. After this five times, the user has to wait 24 hours. In order to accomplish that I raise a counter in my MYSQL database, right next to the user. So where the name of the user is, there is where the counter gets raised. When it reaches 5 it should stop counting and reset after 24 hours. In order to check the time I use a timestamp. I check if the timestamp is older than 24 hours. If that is the case, the counter gets reseted (-5) and the user can comment again. In order to do that, I use the following code, however it never stops at five, my guess is that my comparison is wrong somehow: $counter = "SELECT FROM table VALUES CommentCounterReset WHERE Name = '$name'"; if(!isset($_SESSION['ts'])); { $_SESSION['ts'] = time(); } if ($counter >= 5) { if (time() - $_SESSION['ts'] <= 60*60*24){ echo "You already wrote five comments."; } else { $sql = "UPDATE table SET CommentCounterReset = CommentCounterReset-5 WHERE Name = '$name'"; } } else { $sql = "UPDATE table SET CommentCounterReset = CommentCounterReset+1 WHERE Name = '$name'"; echo "Your comment has been added."; }

  Read the article

• File sorting, Comparison method violates its general contract

  - by user2677383

  My file sorting comparator sometimes invoke java.lang.IllegalArgumentException in android app. I cannot figure out the reason. could you explain that? here is my code block: Comparator cc = new Comparator<File>() { @Override public int compare(File f1, File f2) { if (f1.isDirectory() && !f2.isDirectory()) { return 1; } if (f2.isDirectory() && !f1.isDirectory()) { return -1; } if ((f1.isFile() && f2.isFile()) || (f1.isDirectory() && f2.isDirectory())) { if (sort == 1) { return Util.compareFileName(f2.getName(), f1.getName()); } else if (sort == 2) { return (int) (f1.lastModified() - f2.lastModified()); } else if (sort == 3) { return (int) (f2.lastModified() - f1.lastModified()); } else if (sort == 4) { return (int) (f1.length() - f2.length()); } else if (sort == 5) { return (int) (f2.length() - f1.length()); } else return Util.compareFileName(f1.getName(), f2.getName()); } return f1.compareTo(f2); } }; Util.compareFileName is as followings: public static int compareFileName(String s1, String s2) { int thisMarker = 0; int thatMarker = 0; int s1Length = s1.length(); int s2Length = s2.length(); while (thisMarker < s1Length && thatMarker < s2Length) { String thisChunk = getChunk(s1, s1Length, thisMarker); thisMarker += thisChunk.length(); String thatChunk = getChunk(s2, s2Length, thatMarker); thatMarker += thatChunk.length(); // If both chunks contain numeric characters, sort them numerically int result = 0; if (isDigit(thisChunk.charAt(0)) && isDigit(thatChunk.charAt(0))) { // Simple chunk comparison by length. int thisChunkLength = thisChunk.length(); result = thisChunkLength - thatChunk.length(); // If equal, the first different number counts if (result == 0) { for (int i = 0; i < thisChunkLength; i++) { result = thisChunk.charAt(i) - thatChunk.charAt(i); if (result != 0) { return result; } } } } else { result = thisChunk.compareTo(thatChunk); } if (result != 0) return result; } return s1Length - s2Length; }

  Read the article

• Deriving an HTMLElement Object from jQuery Object

  - by Jasconius

  I'm doing a fairly exhaustive series of DOM manipulations where a few elements (specifically form elements) have some events. I am dynamically creating (actually cloning from a source element) several boxes and assigning a change() event to them. The change event executes, and within the context of the event, "this" is the HTML Element Object. What I need to do at this point however is determine a contact for this HTML Element Object. I have these objects stored already as jQuery entities in assorted arrays, but obviously [HTMLElement Object] != [Object Object] And the trick is that I cannot cast $(this) and make a valid comparison since that would create a new object and the pointer would be different. So... I've been banging my head against this for a while. In the past I've been able to circumvent this problem by doing an innerHTML comparison, but in this case the objects I am comparing are 100% identical, just there's lots of them. Therefore I need a solid comparison. This would be easy if I could somehow derive the HTMLElement object from my originating jQuery object. Thoughts, other ideas? Help. :(

  Read the article

• Warning: comparison with string literals results in unspecified behaviour

  - by nunos

  So I starting the project of writing a simplified sheel for linux in c. I am not at all proficient with c nor with Linux that's exactly the reason I decided it would be a good idea. Starting with the parser, I have already encountered some problems. The code should be straightforward that's why I didn't include any comments. I am getting a warning with gcc: "comparison with string literals results in unspecified behaviour" at the lines commented with "WARNING HERE" (see code below). I have no idea why this causes an warning, but the real problem is that even though I am comparing an "<" to an "<" is doesn't get inside the if... I am looking for an answer for the problem explained, however if there's something that you see in the code that should be improved please say so. Just take in mind I am not that proficient and that this is still a work in progress (or better yet, a work in start). Thanks in advance. #include <stdio.h> #include <unistd.h> #include <string.h> typedef enum {false, true} bool; typedef struct { char **arg; char *infile; char *outfile; int background; } Command_Info; int parse_cmd(char *cmd_line, Command_Info *cmd_info) { char *arg; char *args[100]; int i = 0; arg = strtok(cmd_line, " \n"); while (arg != NULL) { args[i] = arg; arg = strtok(NULL, " \n"); i++; } int num_elems = i; cmd_info->infile = NULL; cmd_info->outfile = NULL; cmd_info->background = 0; int iarg = 0; for (i = 0; i < num_elems; i++) { if (args[i] == "&") //WARNING HERE return -1; else if (args[i] == "<") //WARNING HERE if (args[i+1] != NULL) cmd_info->infile = args[i+1]; else return -1; else if (args[i] == ">") //WARNING HERE if (args[i+1] != NULL) cmd_info->outfile = args[i+1]; else return -1; else cmd_info->arg[iarg++] = args[i]; } cmd_info->arg[iarg] = NULL; return 0; } void print_cmd(Command_Info *cmd_info) { int i; for (i = 0; cmd_info->arg[i] != NULL; i++) printf("arg[%d]=\"%s\"\n", i, cmd_info->arg[i]); printf("arg[%d]=\"%s\"\n", i, cmd_info->arg[i]); printf("infile=\"%s\"\n", cmd_info->infile); printf("outfile=\"%s\"\n", cmd_info->outfile); printf("background=\"%d\"\n", cmd_info->background); } int main(int argc, char* argv[]) { char cmd_line[100]; Command_Info cmd_info; printf(">>> "); fgets(cmd_line, 100, stdin); parse_cmd(cmd_line, &cmd_info); print_cmd(&cmd_info); return 0; }

  Read the article

• Algorithm detect repeating/similiar strings in a corpus of data -- say email subjects, in Python

  - by RizwanK

  I'm downloading a long list of my email subject lines , with the intent of finding email lists that I was a member of years ago, and would want to purge them from my Gmail account (which is getting pretty slow.) I'm specifically thinking of newsletters that often come from the same address, and repeat the product/service/group's name in the subject. I'm aware that I could search/sort by the common occurrence of items from a particular email address (and I intend to), but I'd like to correlate that data with repeating subject lines.... Now, many subject lines would fail a string match, but "Google Friends : Our latest news" "Google Friends : What we're doing today" are more similar to each other than a random subject line, as is: "Virgin Airlines has a great sale today" "Take a flight with Virgin Airlines" So -- how can I start to automagically extract trends/examples of strings that may be more similar. Approaches I've considered and discarded ('because there must be some better way'): Extracting all the possible substrings and ordering them by how often they show up, and manually selecting relevant ones Stripping off the first word or two and then count the occurrence of each sub string Comparing Levenshtein distance between entries Some sort of string similarity index ... Most of these were rejected for massive inefficiency or likelyhood of a vast amount of manual intervention required. I guess I need some sort of fuzzy string matching..? In the end, I can think of kludgy ways of doing this, but I'm looking for something more generic so I've added to my set of tools rather than special casing for this data set. After this, I'd be matching the occurring of particular subject strings with 'From' addresses - I'm not sure if there's a good way of building a data structure that represents how likely/not two messages are part of the 'same email list' or by filtering all my email subjects/from addresses into pools of likely 'related' emails and not -- but that's a problem to solve after this one. Any guidance would be appreciated.

  Read the article

• Which e-reader for O'Reilly books?

  - by Tim

  Most of my programming books are from O'Reilly. Several of them need to be updated, and I'm thinking of going for an e-reader this time. O'Reilly sell their e-books as a multi-format bundle (PDF, EPUB and Mobipocket (Kindle)). Has anyone done a comparison of the formats as for suitability on a small e-reader screen? In Europe, the Kindle DX is not (yet?) available, so I'm not sure whether a Kindle or a Sony (or some other reader) is best suited for these books. Disregarding other advantages/disadvantages of the different readers for now, which one in your opinion gives the best experience with O'Reilly books in particular (or programming books in general)?

  Read the article

• Which one is better deal: iPed or Apple iPad?

  - by Evan Carroll

  The Chinese knockoff iPed is now in the wild and receiving a lot of attention? Does anyone know how the Android distribution it ships with compares with Apple iPad's software pack? How does the look-and-feel of the iPed fit with the Apple iPad? Quite frankly, I have little interest in Apple products because of the company policy. But, cheaper comparable hardware running open-source software is pretty interesting... Anything useful to add.. Here are the Apple iPad specs for comparison. The iPed is priced at $199.00, the Apple iPad starts at $499.00 .

  Read the article

• compareTo() method java is acting weird

  - by Ron Paul

  hi im having trouble getting this to work im getting an error here with my object comparison...how could I cast the inches to a string ( i never used compare to with anything other than strings) , or use comparison operators to compare the intigers, Object comparison = this.inches.compareTo(obj.inches); here is my code so far import java.io.*; import java.util.*; import java.lang.Integer; import java.lang.reflect.Array; public class Distance implements Comparable<Distance> { private static final String HashCodeUtil = null; private int feet; private int inches; private final int DEFAULT_FT = 1; private final int DEFAULT_IN = 1; public Distance(){ feet = DEFAULT_FT; inches = DEFAULT_IN; } public Distance(int ft, int in){ feet = ft; inches = in; } public void setFeet(int ft){ try { if(ft<0){ throw new CustomException("Distance is not negative"); } } catch(CustomException c){ System.err.println(c); feet =ft; } } public int getFeet(){ return feet; } public void setInches(int in){ try { if (in<0) throw new CustomException("Distance is not negative"); //inches = in; } catch(CustomException c) { System.err.println(c); inches = in; } } public int getInches(){ return inches; } public String toString (){ return "<" + feet + ":" + inches + ">"; } public Distance add(Distance m){ Distance n = new Distance(); n.inches = this.inches + m.inches; n.feet = this.feet + m.feet; while(n.inches>12){ n.inches = n.inches - 12; n.feet++; } return n; } public Distance subtract(Distance f){ Distance m = new Distance(); m.inches = this.inches - f.inches; m.feet = this.feet - f.feet; while(m.inches<0){ m.inches = m.inches - 12; feet--; } return m; } @Override public int compareTo(Distance obj) { // TODO Auto-generated method stub final int BEFORE = -1; final int EQUAL = 0; final int AFTER = 1; if (this == obj) return EQUAL; if(this.DEFAULT_IN < obj.DEFAULT_FT) return BEFORE; if(this.DEFAULT_IN > obj.DEFAULT_FT) return AFTER; Object comparison = this.inches.compareTo(obj.inches); if (this.inches == obj.inches) return compareTo(null); assert this.equals(obj) : "compareTo inconsistent with equals"; return EQUAL; } @Override public boolean equals( Object obj){ if (obj != null) return false; if (!(obj intanceof Distance)) return false; Distance that = (Distance)obj; ( this.feet == that.feet && this.inches == that.inches); return true; else return false; } @Override public int hashCode(int, int) { int result = HashCodeUtil.inches; result = HashCodeUtil.hash(result, inches ); result = HashCodeUtil.hash(result, feet); ruturn result; }

  Read the article

• Simplifying a four-dimensional rule table in Matlab: addressing rows and columns of each dimension

  - by Cate

  Hi all. I'm currently trying to automatically generate a set of fuzzy rules for a set of observations which contain four values for each observation, where each observation will correspond to a state (a good example is with Fisher's Iris Data). In Matlab I am creating a four dimensional rule table where a single cell (a,b,c,d) will contain the corresponding state. To reduce the table I am following the Hong and Lee method of row and column similarity checking but I am having difficulty understanding how to address the third and fourth dimensions' rows and columns. From the method it is my understanding that each dimension is addressed individually and if the rule is true, the table is simplified. The rules for merging are as follows: If all cells in adjacent columns or rows are the same. If two cells are the same or if either of them is empty in adjacent columns or rows and at least one cell in both is not empty. If all cells in a column or row are empty and if cells in its two adjacent columns or rows are the same, merge the three. If all cells in a column or row are empty and if cells in its two adjacent columns or rows are the same or either of them is empty, merge the three. If all cells in a column or row are empty and if all the non-empty cells in the column or row to its left have the same region, and all the non-empty cells in the column or row to its right have the same region, but one different from the previously mentioned region, merge these three columns into two parts. Now for the confusing bit. Simply checking if the entire row/column is the same as the adjacent (rule 1) seems simple enough: if (a,:,:,:) == (a+1,:,:,:) (:,b,:,:) == (:,b+1,:,:) (:,:,c,:) == (:,:,c+1,:) (:,:,:,d) == (:,:,:,d+1) is this correct? but to check if the elements in the row/column match, or either is zero (rules 2 and 4), I am a bit lost. Would it be something along these lines: for a = 1:20 for i = 1:length(b) if (a+1,i,:,:) == (a,i,:,:) ... else if (a+1,i,:,:) == 0 ... else if (a,i,:,:) == 0 etc. and for the third and fourth dimensions: for c = 1:20 for i = 1:length(a) if (i,:,c,:) == (i,:,c+1,:) ... else if (i,:,c+1,:) == 0 ... else if (i,:,c,:) == 0 etc. for d = 1:20 for i = 1:length(a) if (i,:,:,d) == (i,:,:,d+1) ... else if (i,:,:,d+1) == 0 ... else if (i,:,:,d) == 0 etc. even any help with four dimensional arrays would be useful as I'm so confused by the thought of more than three! I would advise you look at the paper to understand my meaning - they themselves have used the Iris data but only given an example with a 2D table. Thanks in advance, hopefully!

  Read the article

• Core i7-620M vs Core i5-540M

  - by Shalan

  Hi, I'm recommending a laptop to a colleague, and the specific laptop he has chosen has the above CPU chips as options. Both chips have 2-cores/4-threads. The i7-620M is a 2.66 GHz (4MB Cache) while the i5-540M is a 2.53 GHz (3MB Cache)....both Arrandale architecture. He is a .NET programmer working with SQL Server and Oracle, and occasionally uses Adobe Fireworks for web-related design elements. He also loves playing around in Adobe Premiere Pro, and does a lot of media/video work. Would you notice any significant performance difference between the 2? The laptop manufacturer claims that the battery life on both is the same irrespective of the chip used (although I find that hard to believe), but there is a major cost difference between them, with the Core i7-620M being the more expensive. According to http://ark.intel.com, the one thing that seems different (besides the obvious speeds/frequencies/etc) is a feature called "Embedded" - what is this exactly? You can see the quick comparison here - http://ark.intel.com/Compare.aspx?ids=43544,43560 I would sincerely appreciate any advice me on this. THANKS!

  Read the article

• Java code optimization leads to numerical inaccuracies and errors

  - by rano

  I'm trying to implement a version of the Fuzzy C-Means algorithm in Java and I'm trying to do some optimization by computing just once everything that can be computed just once. This is an iterative algorithm and regarding the updating of a matrix, the clusters x pixels membership matrix U, this is the update rule I want to optimize: where the x are the element of a matrix X (pixels x features) and v belongs to the matrix V (clusters x features). And m is a parameter that ranges from 1.1 to infinity. The distance used is the euclidean norm. If I had to implement this formula in a banal way I'd do: for(int i = 0; i < X.length; i++) { int count = 0; for(int j = 0; j < V.length; j++) { double num = D[i][j]; double sumTerms = 0; for(int k = 0; k < V.length; k++) { double thisDistance = D[i][k]; sumTerms += Math.pow(num / thisDistance, (1.0 / (m - 1.0))); } U[i][j] = (float) (1f / sumTerms); } } In this way some optimization is already done, I precomputed all the possible squared distances between X and V and stored them in a matrix D but that is not enough, since I'm cycling througn the elements of V two times resulting in two nested loops. Looking at the formula the numerator of the fraction is independent of the sum so I can compute numerator and denominator independently and the denominator can be computed just once for each pixel. So I came to a solution like this: int nClusters = V.length; double exp = (1.0 / (m - 1.0)); for(int i = 0; i < X.length; i++) { int count = 0; for(int j = 0; j < nClusters; j++) { double distance = D[i][j]; double denominator = D[i][nClusters]; double numerator = Math.pow(distance, exp); U[i][j] = (float) (1f / (numerator * denominator)); } } Where I precomputed the denominator into an additional column of the matrix D while I was computing the distances: for (int i = 0; i < X.length; i++) { for (int j = 0; j < V.length; j++) { double sum = 0; for (int k = 0; k < nDims; k++) { final double d = X[i][k] - V[j][k]; sum += d * d; } D[i][j] = sum; D[i][B.length] += Math.pow(1 / D[i][j], exp); } } By doing so I encounter numerical differences between the 'banal' computation and the second one that leads to different numerical value in U (not in the first iterates but soon enough). I guess that the problem is that exponentiate very small numbers to high values (the elements of U can range from 0.0 to 1.0 and exp , for m = 1.1, is 10) leads to ver y small values, whereas by dividing the numerator and the denominator and THEN exponentiating the result seems to be better numerically. The problem is it involves much more operations. Am I doing something wrong? Is there a possible solution to get both the code optimized and numerically stable? Any suggestion or criticism will be appreciated.

  Read the article

< Previous Page | 7 8 9 10 11 12 13 14 15 16 17 18  | Next Page >