Umbu and Threads versions are now available and new flags #56
Conversation
| generateChunkOutTif(input, criteriaName, chunk, values, sumAccum); | ||
| } | ||
|
|
||
| TIFFClose(input); |
There was a problem hiding this comment.
@FMota0 could we delete below-commented code?
| @@ -0,0 +1,19 @@ | |||
| #!/bin/bash | |||
|
|
|||
There was a problem hiding this comment.
caja/playground/src/umbu.cpp
Outdated
| using MaskMatrixLine = vector<bool>; | ||
|
|
||
| using MaskMatrix = vector<MaskMatrixLine>; | ||
|
|
| vector<string> convertToVector(int argc, char * argv[]){ | ||
| vector<string> args; | ||
| for(int i = 1; i < argc; i++) args.push_back(argv[i]); | ||
| return args; | ||
| } | ||
|
|
||
| /** |
There was a problem hiding this comment.
there is no need to use two lines for this comment
There was a problem hiding this comment.
... use a single line, break it when it reaches ~80 chars, e.g
| @@ -15,6 +15,9 @@ | |||
| #include "functions/optimized/linear_with_indifference_function.h" | |||
| #include <string> | |||
|
|
|||
| /** | |||
| * Ugly structure to make possible vanilla/optimized have the same input reader | |||
There was a problem hiding this comment.
remove ugly. if you think it should be refactored later, add a TODO
| /** | ||
| ./run -um AEMMF.tif 0.47 -chunk=1111 -ismax -type=linear (args...) | ||
| **/ | ||
|
|
There was a problem hiding this comment.
need improvement. one need to know the all the possible args (I accept adding a TODO telling so)
| void PrometheeUmbu::init(vector<string> args, int divideBy){ | ||
| this->divideBy = divideBy; | ||
|
|
||
| // Get type of function |
There was a problem hiding this comment.
there is no need for this comment. code is simple enough
| cerr << "Error: incorrect arguments." << endl; | ||
| exit(0); | ||
| } | ||
|
|
| } | ||
| } | ||
|
|
||
| /** |
There was a problem hiding this comment.
it could be improved. describe the purpose of the function and its parameters (all)
adopt the following standard
paramentername - description
paramentername - description
| * input, values, sumAccum, cnt are arguments to make calc of flow | ||
| * */ | ||
| void PrometheeUmbu::generateChunkOutTifUnbu(string &outputFile, string &nextFile, TIFF *input, vector<ldouble> & values, vector<ldouble> & sumAccum, vector<unsigned int> &cntAccum){ | ||
|
|
There was a problem hiding this comment.
it is not clear why we need current out and next out. Unclear code should be explained
| TIFF *out = TIFFOpen(outputFile.c_str(), "rm"); | ||
| TIFF *nxt = openFile(nextFile, this->width, this->height); | ||
|
|
||
| ldouble *line = new ldouble[this->width]; |
There was a problem hiding this comment.
it seems line should be inputline
| ldouble *line = new ldouble[this->width]; | ||
| ldouble *outline = new ldouble[this->width]; | ||
| for (int i = 0; i < this->height; i++){ | ||
|
|
There was a problem hiding this comment.
out of curiosity, is this read operation buffered? and, for our typical tiffs, how big is a line (in bytes)
The flag -hq now is available you can use it like
./run samples/sample3/input samples/sample3/meta out/sample3 -hq=2also, the flag of vanilla was changed, to use vanilla you can run as in the example:
./run -van samples/sample1/input samples/sample1/meta out/sample1 -hq=4And the main is simpler now.
The flag -thr will run promethee optimized with threads.
The flag -um will run promethee umbu.
For promethee umbu the following command options should be specified:
Example:
./run -um tif weight -chunk=1111 -ismax -type=linear (args...)-chunk=, for max amount of distinct values in a chunk.
-ismax, for indicate that the objective is to maximaze.
-type=, what type of function to be used ("linear" or "linearWithIndifference")
(args...), represent the arguments for the specified type of function
Solve issues #45 and #50.
The issue #37 can be seen as solved, since the input reader change depending of the promethee version, Umbu reads from tif, others read from txt