19 |
20 |
21 | function varargout=ann2rr(varargin)
22 |
23 | [RR,tms]=ann2rr(recordName,annotator,N,N0,consecutiveOnly)
24 |
25 | Wrapper to WFDB ANN2RR:
26 | http://www.physionet.org/physiotools/wag/ann2rr-1.htm
27 |
28 | Reads a WFDB record and Annotation file to return:
29 |
30 |
31 | RR
32 | Nx1 vector of integers representing the duration of the RR
33 | interval in samples.
34 |
35 | tms
36 | Nx1 vector of integers representing the begining of the RR
37 | interval in samples.
38 |
39 | Required Parameters:
40 |
41 | recorName
42 | String specifying the name of the record in the WFDB path or
43 | in the current directory.
44 |
45 | annotator -
46 | String specifying the name of the annotation file in the WFDB path or
47 | in the current directory.
48 |
49 | Optional Parameters are:
50 |
51 | N
52 | A 1x1 integer specifying the sample number at which to stop reading the
53 | record file (default read all = N).
54 | N0
55 | A 1x1 integer specifying the sample number at which to start reading the
56 | annotion file (default 1 = begining of the record).
57 |
58 | consecutiveOnly
59 | A 1x1 boolean. If true, prints intervals between consecutive valid
60 | annotaions only (default =true).
61 |
62 |
63 | Written by Ikaro Silva, 2013
64 | Last Modified: January, 16, 2013
65 | Version 1.1
66 |
67 | Since 0.0.1
68 | %Example
69 | [rr,tm]=ann2rr('challenge/2013/set-a/a01','fqrs');
70 |
19 |
20 |
21 | function varargout=getWfdbClass(varargin)
22 |
23 | wfdbClass=getWfdbClass(comandName)
24 |
25 | Returns a 'wfdbClass' Java object defined my the string 'commandName' with system
26 | wide run time settings defined by the toolbox WFDBLOALIB file. This class will
27 | execute the WFDB native binary associate with 'commandName'.
28 |
29 | Written by Ikaro Silva, November 23, 2013
30 | Last Modified: January 16, 2014
31 |
32 | Since 0.9.5
33 | See also WFDBEXEC, WFDB, WFDBLOADLIB
34 |
19 |
20 |
21 | function varargout=snip(varargin)
22 |
23 | err=snip(inputRecord,outputRecord,beginTime,stopTime,inputAnn,outFormat)
24 |
25 | Wrapper to WFDB SNIP:
26 | http://www.physionet.org/physiotools/wag/snip-1.htm
27 |
28 | Copy an excerpt of a WFDB record
29 |
30 |
31 | Input Parameters:
32 | inputRecord
33 | String specifying the input WFDB record file.
34 |
35 | outputRecord
36 | String specifying the output WFDB record file name that will be generated.
37 |
38 | beginTime (Optional)
39 | Integer specifying start time of the output WFDB record. Default is
40 | the beginning of the input record.
41 |
42 | stopTime (Optional)
43 | Integer specifying end time of the output WFDB record. Defaut is
44 | end of input record.
45 |
46 | inputAnn (Optional)
47 | String specifying the annotation files to convert along with the
48 | given record. Defaults i none (empty).
49 |
50 | outFormat (Optional)
51 | String specifying the output format (see http://www.physionet.org/physiotools/wag/header-5.htm).
52 | Default is the same as input record.
53 |
54 | Output Parameters:
55 | err (Optional)
56 | String spefiying any error messages. If empty, conversion was
57 | sucessfull.
58 |
59 | Written by Ikaro Silva, 2015
60 | Last Modified: -
61 | Version 1.0
62 | Since 0.9.10
63 |
64 |
65 | %Example- Generate a record from the first minute of mitdb/100
66 | Fs=360;
67 | err=snip('mitdb/100','100cut',[],Fs*60);
68 | [sig2,Fs,tm1]=rdsamp('mitdb/100');
69 | [sig2,Fs,tm2]=rdsamp('100cut');
70 | plot(tm1,sig1(:,1));hold on;grid on
71 | plot(tm2,sig2(:,1),'r')
72 |
73 |
74 | See also RDSAMP, RDANN, WFDBDESC
75 |
19 |
20 |
21 | function varargout=sortann(varargin)
22 |
23 | sortann(recName,annName,beginTime,stopTime,outFile)
24 |
25 | Wrapper to WFDB SORTANN:
26 | http://www.physionet.org/physiotools/wag/sortan-1.htm
27 |
28 | Rewrites the annotation file specified by recName and annName, arranging its contents in
29 | canonical (time, num, and chan) order. The sorted (output) annotation file is always written
30 | to the current directory. If the input annotation file is in the current directory, SORTANN
31 | replaces it unless you specify a different output annotator name (using the outFile option).
32 | If the input annotations are already in the correct order, no output is written unless you
33 | have used the outFile option.
34 |
35 |
36 | Input Parameters:
37 | recName
38 | String specifying the WFDB record file.
39 |
40 | annName
41 | String specifying the reference WFDB annotation file.
42 |
43 | stopTime (Optional)
44 | String specifying the start time in WFDB format (default is beginning of
45 | record).
46 |
47 | stopTime (Optional)
48 | String specifying the stop time in WFDB format (default is end of
49 | record).
50 |
51 | outFile (Optional)
52 | String specifying the output annotation file name.
53 |
54 |
55 | Written by Ikaro Silva, 2013
56 | Last Modified: -
57 | Version 1.0
58 | Since 0.9.0
59 |
60 | %Example (this will generate a /mitdb/100.sortedATR file at your directory):
61 |
62 | sortann('mitdb/100','atr',[],[],'sortedATR');
63 | ann=rdann('mitdb/100','sortedATR');
64 |
65 |
66 | See also RDANN
67 |
19 |
20 |
21 | function varargout=sumann(varargin)
22 |
23 | report=sumann(recName,annName,stopTime,qrsAnnotationsOnly)
24 |
25 | Wrapper to WFDB SUMANN:
26 | http://www.physionet.org/physiotools/wag/sumann-1.htm
27 |
28 | Reads a WFDB annotation file and summarize its contents.
29 |
30 | Ouput Parameters:
31 |
32 | report
33 | String with the contaning summary of the contents, including the
34 | number of annotations of each type as well the duration and number of
35 | episodes of each rhythm and signal quality.
36 |
37 | Input Parameters:
38 | recName
39 | String specifying the WFDB record file.
40 |
41 | annName
42 | String specifying the reference WFDB annotation file.
43 |
44 | stopTime (Optional)
45 | String specifying the stop time in WFDB format (default is end of
46 | record).
47 |
48 | qrsAnnotationsOnly (Optional)
49 | 1x1 Boolean. If true, summarize QRS annotation only (default = 0).
50 |
51 |
52 | Written by Ikaro Silva, 2013
53 | Last Modified: -
54 | Version 1.0
55 | Since 0.9.0
56 |
57 | %Example (this will generate a /mitdb/100.qrs file at your directory):
58 |
59 | report=sumann('mitdb/100','atr');
60 |
61 |
62 |
63 | See also RDANN, MXM, WFDBTIME, BXB
64 |
19 |
20 |
21 | function varargout=tach(varargin)
22 |
23 | [hr]=tach(recordName,annotator,N,N0,ouputSize)
24 |
25 | Wrapper to WFDB TACH:
26 | http://www.physionet.org/physiotools/wag/tach-1.htm
27 |
28 | Reads a WFDB record and Annotation file to return:
29 |
30 |
31 | hr
32 | Nx1 vector of doubles representing a uniformly sampled and
33 | smoothed instantaneous heart rate signal. The output are samples
34 | of the instantaneous heart rate signal in units of beats per minute.
35 |
36 | Required Parameters:
37 |
38 | recorName
39 | String specifying the name of the record in the WFDB path or
40 | in the current directory.
41 |
42 | annotator -
43 | String specifying the name of the annotation file in the WFDB path or
44 | in the current directory.
45 |
46 | Optional Parameters are:
47 |
48 | N
49 | A 1x1 integer specifying the sample number at which to stop reading the
50 | annotation file (default read all = N).
51 | N0
52 | A 1x1 integer specifying the sample number at which to start reading the
53 | annotation file (default 1 = begining of the record).
54 |
55 | outputSize
56 |
57 | A 1x1 integer specifying the number of output samples (ie estimated
58 | heart rate intervals) such that the output 'hr' is a vector of
59 | size (outputSize-1) x 1.
60 |
61 |
62 | Written by Ikaro Silva, 2013
63 | Last Modified: January 24, 2014
64 | Version 1.1
65 |
66 | Since 0.0.1
67 |
68 | %Example 1- Read a signal and annotaion from PhysioNet's Remote server:
69 | [hr]=tach('challenge/2013/set-a/a01','fqrs');
70 | plot(hr);grid on;hold on
71 |
19 |
20 |
21 | function wfdb2mat(varargin)
22 |
23 | wfdm2mat(recordName,signaList,N,N0)
24 |
25 | Wrapper to WFDB WFDB2MAT:
26 | http://physionet.org/physiotools/wag/wfdb2m-1.htm
27 |
28 | Converts a WFDB-compatible signal file to MATLAB/Octave *.mat file.
29 | The output files are recordNamem.mat and recordNamem.hea. The standard output of
30 | WFDB2MAT will be saved in a file recordNamem.info.
31 |
32 | Required Parameters:
33 |
34 | recorName
35 | String specifying the name of the record in the WFDB path or
36 | in the current directory.
37 |
38 | Optional Parameters are:
39 |
40 | signalList
41 | A Mx1 array of integers. Read only the signals (columns)
42 | named in the signalList (default: read all signals).
43 | N
44 | A 1x1 integer specifying the sample number at which to stop reading the
45 | record file (default read all the samples = N).
46 | N0
47 | A 1x1 integer specifying the sample number at which to start reading the
48 | record file (default 1 = first sample).
49 |
50 |
51 |
52 | NOTE:
53 | You can use the WFDB2MAT command in order to convert the record data into a *.mat file,
54 | which can then be loaded into MATLAB/Octave's workspace using the LOAD command.
55 | This will load the signal data in raw units (use RDMAT to load the signal in physical units).
56 |
57 |
58 | Written by Ikaro Silva, 2014
59 | Last Modified: September 15, 2014
60 | Version 0.1
61 |
62 | Since 0.9.7
63 |
64 | %Example:
65 | wfdb2mat('mitdb/200')
66 | [tm,signal,Fs,labels]=rdmat('200m');
67 |
68 |
69 | See also RDSAMP, RDMAT
70 |
19 |
20 |
21 | function varargout=wfdbdownload(varargin)
22 |
23 | [success,files_saved]=wfdbdownload(recordName)
24 |
25 | Downloads a WFDB record with recordName
26 | and associated files from PhysioNet server and store is on the WFB Toolbox
27 | cache directory.
28 |
29 | The toolbox cache directory is determined by the foolowing toolbox
30 | configuation parameters obtained by running:
31 |
32 | [~,config]=wfdbloadlib;
33 |
34 | config.CACHE -Boolean. If true this wfdbdownlaod will attempt to
35 | download record
36 |
37 | config.CACHE_DEST -Destion of the cached files on the user's system.
38 | It shoudl be safe to delete the cached files, they
39 | can be re-obtained when CACHE==1.
40 |
41 | config.CACHE_SOURCE -Source of the cached files (default is PhysioNet's
42 | server at physionet.org/physiobank/database/
43 |
44 |
45 | Optional output parameters:
46 |
47 | success
48 | Integer. If 0, could not download files, if -1, file already
49 | exists of CACHE==0. If success>0, an integer representing the number of files
50 | downloaded.
51 |
52 | files_saved
53 | A cell array of string specifying the saved files full path.
54 |
55 |
56 | Written by Ikaro Silva, April 6, 2015
57 | Last Modified: -
58 | Version 0.1
59 |
60 | Since 0.0.1
61 | %Example:
62 | [success,files_saved]=wfdbdownload('mitdb/102')
63 |
64 |
65 | See also WFDBLOADLIB, RDSAMP
66 |
19 |
20 |
21 | function [varargout]=wfdbloadlib(varargin)
22 |
23 | [isloaded,config]=wfdbloadlib(debugLevel,networkWaitTime)
24 |
25 | Loads the WDFDB libarary if it has not been loaded already into the
26 | MATLAB classpath. And optionally prints configuration environment and debug information
27 | regarding the settings used by the classes in the JAR file.
28 |
29 | Inputs:
30 |
31 | debugLevel
32 | (Optional) 1x1 integer between 0 and 5 represeting the level of debug information to output from
33 | Java class when output configuration information. Level 0 (no debug information),
34 | level =5 is maximum level of information output by the class (logger set to finest). Default is level 0.
35 |
36 | networkWaitTime
37 | (Optional) 1x1 integer representing the longest time in
38 | milliseconds for which the JVM should wait for a data stream from
39 | PhysioNet (default is =1000 , ie one second). If you need to change this time to a
40 | longer value across the entire toolbox, it is better modify to default value in the source
41 | code below and restart MATLAB.
42 |
43 |
44 | Written by Ikaro Silva, 2013
45 | Last Modified: April 7, 2015
46 | Since 0.0.1
47 |
19 |
20 |
21 | function wfdbtest(varargin)
22 | This script will test the installation of the WFDB Application Toolbox
23 |
24 | Written by Ikaro Silva, 2013
25 |
26 | Last Modified: October 15, 2014
27 |
28 | Version 1.2
29 | Since 0.0.1
30 |
31 | See also wfdb, rdsamp
32 |
19 |
20 |
21 | function varargout=wfdbtime(varargin)
22 |
23 | [timeStamp,dateStamp]=wfdbtime(recordName,samples)
24 |
25 | Wrapper to WFDB WFDBTIME:
26 | http://www.physionet.org/physiotools/wag/wfdbti-1.htm
27 |
28 | Converts sample indices from recordName into timeStamp and dateStamps.
29 | Returns:
30 |
31 | timesStamp
32 | Nx1 vector of cell Strings representing times stamps with respect to the
33 | first sample in recordName.
34 |
35 | dateStamp
36 | Nx1 vector of cell Strings representing date stamps with respect to the
37 | first sample in recordName.
38 |
39 |
40 | Required Parameters:
41 |
42 | recorName
43 | String specifying the name of the record in the WFDB path or
44 | in the current directory.
45 |
46 | samples
47 | Nx1 vector of integers (indices) of samples from the signal in recordName (indices are
48 | relative to the first sample).
49 |
50 |
51 | %Example
52 | [timeStamp,dateStamp]=wfdbtime('challenge/2013/set-a/a01',[1 10 30]')
53 |
54 |
55 |
56 | Written by Ikaro Silva, 2013
57 | Last Modified: March 24, 2014
58 | Version 1.1
59 | Since 0.0.1
60 |
61 | See also RDANN, WFDBDESC
62 |
19 |
20 |
21 | function varargout=snip(varargin)
22 |
23 | err=snip(inputRecord,outputRecord,beginTime,stopTime,inputAnn,outSignalList,outFs)
24 |
25 | Wrapper to WFDB SNIP:
26 | http://www.physionet.org/physiotools/wag/snip-1.htm
27 |
28 | Copy an excerpt of a WFDB record
29 |
30 |
31 | Input Parameters:
32 | inputRecord
33 | String specifying the input WFDB record file.
34 |
35 | outputRecord
36 | String specifying the output WFDB record file name that will be generated.
37 |
38 | beginTime (Optional)
39 | Integer specifying start time of the output WFDB record. Default is
40 | the beginning of the input record.
41 |
42 | stopTime (Optional)
43 | Integer specifying end time of the output WFDB record. Defaut is
44 | end of input record.
45 |
46 | inputAnn (Optional)
47 | String specifying the annotation files to convert along with the
48 | given record. Defaults i none (empty).
49 |
50 | outSignalList (Optional)
51 | Array of integers specifying which signals to convert. Default is
52 | to use all signals.
53 |
54 | outFormat (Optional)
55 | String specifying the output format (see http://www.physionet.org/physiotools/wag/header-5.htm).
56 | Default is the same as input record.
57 |
58 | Output Parameters:
59 | err (Optional)
60 | String spefiying any error messages. If empty, conversion was
61 | sucessfull.
62 |
63 | Written by Ikaro Silva, 2015
64 | Last Modified: -
65 | Version 1.0
66 | Since 1.0
67 |
68 | See also RDSAMP, RDANN, WFDBDESC
69 |
70 |
71 |
72 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/.gitignore:
--------------------------------------------------------------------------------
1 | /100m.hea
2 | /100m.mat
3 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/PTBDB.m:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/PTBDB.m
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/README.txt:
--------------------------------------------------------------------------------
1 |
2 | -------To install the WFDB Application Toolbox:-------
3 |
4 | 1) Unzip the zip file into the directory you wish to install the toolbox
5 |
6 |
7 | If you run into any problems and need to contact us,
8 | please send the entire output of this script.
9 |
10 | 2) From within MATLAB, cd into the directory and add it to your path:
11 |
12 | cd wfdb-app-toolbox-x-x-x;cd mcode
13 | addpath(pwd);savepath
14 | wfdbdemo %Optional demoing of the toolbox
15 |
16 |
17 |
18 | -------Getting help and information about the WFDB Toolbox--------
19 | For a information about the Toolbox and the list of functions associated with it
20 | type:
21 |
22 | wfdb
23 |
24 | at the MATLAB prompt.
25 |
26 |
27 |
28 | -------To Uninstall the toolbox:-------
29 |
30 | 1)From MATLAB, find where the toolbox is installed:
31 |
32 | install_dir=which('wfdb')
33 |
34 | 2) Remove the directory from the MATLAB path:
35 | rmpath(install_dir);
36 |
37 | 3)(Optional) Remove the Toolbox files permanently from your machine:
38 | delete(install_dir)
39 |
40 |
41 |
42 | -------CONTACT: For help, feedback, and support please contact us at the community Forum:
43 | https://groups.google.com/forum/#!forum/wfdb-app-toolbox
44 |
45 | *When contacting us about issues with the Toolbox, please send us the output of
46 | the "wfdbtest" script.
47 |
48 |
49 |
50 | -------Join our community!! -------
51 | Join our community at:
52 |
53 | https://groups.google.com/forum/#!forum/wfdb-app-toolbox
54 |
55 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/ann2rr.m:
--------------------------------------------------------------------------------
1 | function varargout=ann2rr(varargin)
2 | %
3 | % [RR,tms]=ann2rr(recordName,annotator,N,N0,consecutiveOnly)
4 | %
5 | % Wrapper to WFDB ANN2RR:
6 | % http://www.physionet.org/physiotools/wag/ann2rr-1.htm
7 | %
8 | % Reads a WFDB record and Annotation file to return:
9 | %
10 | %
11 | % RR
12 | % Nx1 vector of integers representing the duration of the RR
13 | % interval in samples.
14 | %
15 | % tms
16 | % Nx1 vector of integers representing the begining of the RR
17 | % interval in samples.
18 | %
19 | % Required Parameters:
20 | %
21 | % recorName
22 | % String specifying the name of the record in the WFDB path or
23 | % in the current directory.
24 | %
25 | % annotator -
26 | % String specifying the name of the annotation file in the WFDB path or
27 | % in the current directory.
28 | %
29 | % Optional Parameters are:
30 | %
31 | % N
32 | % A 1x1 integer specifying the sample number at which to stop reading the
33 | % record file (default read all = N).
34 | % N0
35 | % A 1x1 integer specifying the sample number at which to start reading the
36 | % annotion file (default 1 = begining of the record).
37 | %
38 | % consecutiveOnly
39 | % A 1x1 boolean. If true, prints intervals between consecutive valid
40 | % annotaions only (default =true).
41 | %
42 | %
43 | % Written by Ikaro Silva, 2013
44 | % Last Modified: January, 16, 2013
45 | % Version 1.1
46 | %
47 | % Since 0.0.1
48 | % %Example
49 | %[rr,tm]=ann2rr('challenge/2013/set-a/a01','fqrs');
50 |
51 | %endOfHelp
52 |
53 | persistent javaWfdbExec config
54 | if(isempty(javaWfdbExec))
55 | [javaWfdbExec,config]=getWfdbClass('ann2rr');
56 | end
57 |
58 | %Set default pararamter values
59 | inputs={'recordName','annotator','N','N0','consecutiveOnly'};
60 | outputs={'data(:,2)','data(:,1)'};
61 | N=[];
62 | N0=1;
63 | consecutiveOnly=1;
64 | for n=1:nargin
65 | if(~isempty(varargin{n}))
66 | eval([inputs{n} '=varargin{n};'])
67 | end
68 | end
69 |
70 | N0=num2str(N0-1); %-1 is necessary because WFDB is 0 based indexed.
71 | wfdb_argument={'-r',recordName,'-a',annotator,'-f',['s' N0]};
72 |
73 | if(~isempty(N))
74 | wfdb_argument{end+1}='-t';
75 | %-1 is necessary because WFDB is 0 based indexed.
76 | wfdb_argument{end+1}=['s' num2str(N-1)];
77 | end
78 |
79 | if(consecutiveOnly)
80 | wfdb_argument{end+1}='-c';
81 | end
82 | wfdb_argument{end+1}='-V';
83 |
84 | data=javaWfdbExec.execToDoubleArray(wfdb_argument);
85 | if(config.inOctave)
86 | data=java2mat(data);
87 | end
88 | for n=1:nargout
89 | eval(['varargout{n}=' outputs{n} ';'])
90 | end
91 |
92 |
93 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/convolution_sb.m:
--------------------------------------------------------------------------------
1 |
2 | function SC_Conv=convolution_sb(SC)
3 | %Convolution 3x3
4 | h=[1 0 -1;1 0 -1;1 0 -1] ;
5 | for i=1:size(SC,1)-2
6 | for j=1:size(SC,2)-2
7 | x=SC(i,j)*h(1,1)+SC(i,j+1)*h(1,2)+SC(i,j+2)*h(1,3);
8 | y=SC(i+1,j)*h(2,1)+SC(i+1,j+1)*h(2,2)+SC(i+1,j+2)*h(2,3);
9 | z=SC(i+2,j)*h(3,1)+SC(i+2,j+1)*h(3,2)+SC(i+2,j+2)*h(3,3);
10 | SC_Conv(i,j)=x+y+z;
11 | end
12 | end
13 | end
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/convolution_sb_2.m:
--------------------------------------------------------------------------------
1 |
2 | function SC_Conv=convolution_sb_2(SC)
3 | %Convolution 5x5
4 | h=[1 0 0 0 1;0 1 0 1 0;0 0 1 0 0;0 1 0 1 0;1 0 0 0 1 ] ;
5 | for i=1:size(SC,1)-4
6 | for j=1:size(SC,2)-4
7 | x=SC(i,j)*h(1,1)+SC(i,j+1)*h(1,2)+SC(i,j+2)*h(1,3)+SC(i,j+3)*h(1,4)+SC(i,j+4)*h(1,5);
8 | y=SC(i+1,j)*h(2,1)+SC(i+1,j+1)*h(2,2)+SC(i+1,j+2)*h(2,3)+SC(i+1,j+3)*h(2,4)+SC(i+1,j+4)*h(2,5);
9 | z=SC(i+2,j)*h(3,1)+SC(i+2,j+1)*h(3,2)+SC(i+2,j+2)*h(3,3)+SC(i+2,j+3)*h(3,4)+SC(i+2,j+4)*h(3,5);
10 | w=SC(i+3,j)*h(4,1)+SC(i+3,j+1)*h(4,2)+SC(i+3,j+2)*h(4,3)+SC(i+3,j+3)*h(4,4)+SC(i+3,j+4)*h(4,5);
11 | l=SC(i+4,j)*h(5,1)+SC(i+4,j+1)*h(5,2)+SC(i+4,j+2)*h(5,3)+SC(i+4,j+3)*h(5,4)+SC(i+4,j+4)*h(5,5);
12 | SC_Conv(i,j)=x+y+z+w+l;
13 | end
14 | end
15 | end
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/example/100a.hea:
--------------------------------------------------------------------------------
1 | 100a 0 360 650000
2 | # From mitdb/100
3 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/example/100s.atr:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/example/100s.atr
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/example/100s.dat:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/example/100s.dat
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/example/100s.hea:
--------------------------------------------------------------------------------
1 | 100s 2 360 21600
2 | 100s.dat 212 200 11 1024 995 21537 0 MLII
3 | 100s.dat 212 200 11 1024 1011 -3962 0 V5
4 | # 69 M 1085 1629 x1
5 | # Aldomet, Inderal
6 | #Produced by xform from record 100, beginning at 0:0
7 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/example/a01.dat:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/example/a01.dat
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/example/a01.entry1:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/example/a01.entry1
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/example/a01.fqrs:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/example/a01.fqrs
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/example/a01.hea:
--------------------------------------------------------------------------------
1 | a01 4 1000 60000
2 | a01.dat 16 10/uV 12 0 -33 14459 0 AECG1
3 | a01.dat 16 10/uV 12 0 -67 -21975 0 AECG2
4 | a01.dat 16 10/uV 12 0 30 -14189 0 AECG3
5 | a01.dat 16 10/uV 12 0 -35 -17968 0 AECG4
6 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/example/f1o02.dat:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/example/f1o02.dat
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/example/f1o02.ecg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/example/f1o02.ecg
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/example/f1o02.hea:
--------------------------------------------------------------------------------
1 | f1o02 2 250 1811007
2 | f1o02.dat 16 2000 16 0 16864 26424 0 RESP
3 | f1o02.dat 16 2000 16 0 16376 -31952 0 ECG
4 | # Age: 73 Sex: F
5 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/example/fort.21:
--------------------------------------------------------------------------------
1 | n posicion 100s
2 |
3 | frecuencia de muestreo = 360
4 | posicion en milisegundos
5 |
6 | 1 211.
7 | 2 1025.
8 | 3 1836.
9 | 4 2625.
10 | 5 3416.
11 | 6 4205.
12 | 7 5022.
13 | 8 5675.
14 | 9 6672.
15 | 10 7513.
16 | 11 8322.
17 | 12 9113.
18 | 13 9883.
19 | 14 10725.
20 | 15 11580.
21 | 16 12402.
22 | 17 13230.
23 | 18 14052.
24 | 19 14847.
25 | 20 15644.
26 | 21 16436.
27 | 22 17258.
28 | 23 18127.
29 | 24 18950.
30 | 25 19736.
31 | 26 20527.
32 | 27 21300.
33 | 28 22088.
34 | 29 22902.
35 | 30 23716.
36 | 31 24544.
37 | 32 25388.
38 | 33 26194.
39 | 34 26969.
40 | 35 27769.
41 | 36 28558.
42 | 37 29416.
43 | 38 30258.
44 | 39 31083.
45 | 40 31886.
46 | 41 32722.
47 | 42 33513.
48 | 43 34302.
49 | 44 35122.
50 | 45 35966.
51 | 46 36847.
52 | 47 37669.
53 | 48 38447.
54 | 49 39250.
55 | 50 40058.
56 | 51 40858.
57 | 52 41694.
58 | 53 42525.
59 | 54 43350.
60 | 55 44161.
61 | 56 44950.
62 | 57 45730.
63 | 58 46538.
64 | 59 47380.
65 | 60 48213.
66 | 61 49044.
67 | 62 49850.
68 | 63 50627.
69 | 64 51425.
70 | 65 52205.
71 | 66 52997.
72 | 67 53852.
73 | 68 54700.
74 | 69 55522.
75 | 70 56305.
76 | 71 57091.
77 | 72 57877.
78 | 73 58694.
79 | 74 59505.
80 | RMAX = 1.000 Factor d'escala senyal
81 | PA_RMAX = 167589.703 Factor d'escala senyal passa_alt
82 | PB_RMAX = 271.431 Factor d'escala senyal passa_baix
83 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/example/fort.6:
--------------------------------------------------------------------------------
1 | QRS en el ECG: 75
2 | QRS en la conv.: 74
3 | QRS conformados: 74
4 | .. 72 beats
5 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/example/r:
--------------------------------------------------------------------------------
1 |
2 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/getWfdbClass.m:
--------------------------------------------------------------------------------
1 | function varargout=getWfdbClass(varargin)
2 | %
3 | % wfdbClass=getWfdbClass(comandName)
4 | %
5 | % Returns a 'wfdbClass' Java object defined my the string 'commandName' with system
6 | % wide run time settings defined by the toolbox WFDBLOALIB file. This class will
7 | % execute the WFDB native binary associate with 'commandName'.
8 | %
9 | % Written by Ikaro Silva, November 23, 2013
10 | % Last Modified: January 16, 2014
11 | %
12 | % Since 0.9.5
13 | % See also WFDBEXEC, WFDB, WFDBLOADLIB
14 |
15 | %endOfHelp
16 |
17 | mlock
18 | persistent config
19 | if(isempty(config))
20 | %Add classes to dynamic path
21 | [~,config]=wfdbloadlib;
22 | end
23 |
24 | inputs={'commandName'};
25 | outputs={'javaWfdbExec','config'};
26 | for n=1:nargin
27 | if(~isempty(varargin{n}))
28 | eval([inputs{n} '=varargin{n};'])
29 | end
30 | end
31 |
32 | %Load the Java class in memory if it has not been loaded yet
33 | %with system wide parameters defined by wfdbloadlib.m
34 | javaWfdbExec=javaObject('org.physionet.wfdb.Wfdbexec',commandName,config.WFDB_CUSTOMLIB);
35 | javaWfdbExec.setInitialWaitTime(config.NETWORK_WAIT_TIME);
36 | javaWfdbExec.setLogLevel(config.DEBUG_LEVEL);
37 | javaWfdbExec.setWFDB_PATH(config.WFDB_PATH);
38 | javaWfdbExec.setWFDBCAL(config.WFDBCAL);
39 |
40 | for n=1:nargout
41 | eval(['varargout{n}=' outputs{n} ';'])
42 | end
43 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/info.xml:
--------------------------------------------------------------------------------
1 |
3 |
4 | 2013a
5 | WFDB Toolbox
6 | toolbox
7 |
8 | html
9 | $toolbox/matlab/icons/bookicon.gif
10 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/mrgann.m:
--------------------------------------------------------------------------------
1 | function mrgann(varargin)
2 | %
3 | % mrgann(recName,annName1,annName2,outAnn,verbose)
4 | %
5 | % Wrapper to WFDB MRGANN:
6 | % http://www.physionet.org/physiotools/wag/mrgann-1.htm
7 | %
8 | %
9 | % Reads a pair of annotation files (annName1, annName2) for the specified
10 | % record (recName), and writes a third annotation file (specified by outAnn)
11 | % for the same record. Typical applications of MRGANN include combining annotation
12 | % files that apply to different signals within a multi-signal record, and replacing
13 | % a segment of an annotation file with annotations from another file. MRGANN cannot
14 | % concatenate annotation files from different records (e.g., segments of a multi-segment record).
15 | %
16 | %
17 | %Required Parameters:
18 | %
19 | % recName
20 | % String specifying the name of the record in the WFDB path or
21 | % in the current directory.
22 | %
23 | % annName1
24 | % String specifying the name of the first WFDB annotation file to be
25 | % merged.
26 | %
27 | % annName2
28 | % String specifying the name of the second WFDB annotation file to be
29 | % merged.
30 | %
31 | % outAnn
32 | % String specifying the name of the output WFDB annotation file
33 | % containing the merged annotations.
34 | %
35 | %
36 | % Optional Parameters are:
37 | %
38 | % verbose
39 | % Boolean. If true warns about simultaneous annoations with matching
40 | % chan fields (default = true).
41 | %
42 | %
43 | % MATLAB wrapper Written by Ikaro Silva, 2013
44 | % Last Modified: 6/13/2013
45 | % Version 1.0
46 | %
47 | % Since 0.9.0
48 | %
49 | % See also BXB, RDANN, WRANN
50 | %
51 | %
52 | % %Example 1- Read a signal and annotation from PhysioNet's Remote server:
53 | % %and merge with calculated WRQS annotation
54 | % wqrs('mitdb/100');
55 | % mrgann('mitdb/100','atr','wqrs','testAnn')
56 | %
57 | %
58 | %
59 | % See also wfdbtime, wrann
60 |
61 | %endOfHelp
62 |
63 | persistent javaWfdbExec
64 | if(isempty(javaWfdbExec))
65 | javaWfdbExec=getWfdbClass('mrgann');
66 | end
67 |
68 | %Set default pararamter values
69 | % [ann,type,subtype,chan,num]=rdann(recordName,annotator,C,N,N0)
70 | inputs={'recName','annName1','annName2','outAnn','verbose'};
71 | verbose=1;
72 | for n=1:nargin
73 | if(~isempty(varargin{n}))
74 | eval([inputs{n} '=varargin{n};'])
75 | end
76 | end
77 |
78 | wfdb_argument={'-r',recName,'-i',annName1,annName2,'-o',outAnn};
79 |
80 | if(verbose)
81 | wfdb_argument{end+1}='-v';
82 | end
83 | javaWfdbExec.execToStringList(wfdb_argument);
84 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/pbsearch.m:
--------------------------------------------------------------------------------
1 | function pbsearch
2 | %
3 | % pbsearch()
4 | %
5 | %
6 | % Launches PhsyioNet's record search tool from within
7 | % MATLAB's web browser.
8 | %
9 | %
10 | % Written by Ikaro Silva, 2014
11 | % Last Modified: October 8, 2014
12 | % Version 1.0
13 | %
14 | % Since 0.9.8
15 | %
16 | % %Example - Launch MATLAB webrowser at PhsyioNet's record search tool
17 | % pbsearch
18 | %
19 | % See also WFDBDESC, PHYSIONETDB
20 |
21 | %endOfHelp
22 |
23 |
24 |
25 | web('http://physionet.org/cgi-bin/pbs/pbsearch?subject=&comp_op=&sval=&name_num=&help_on=on&res_action=&sq_action=')
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/pooling_sb.m:
--------------------------------------------------------------------------------
1 | function pooling=pooling_sb(SC)
2 | %pooling 2x2
3 | m=1;
4 | for i=1:2:size(SC,1)-1
5 | n=1;
6 | for j=1:2:size(SC,2)-1
7 | pooling(m,n)=max(max([SC(i,j) SC(i,j+1);SC(i+1,j) SC(i+1,j+1)]));
8 | n=n+1;
9 | end
10 | m=m+1;
11 | end
12 | end
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/ptbdb_control.txt:
--------------------------------------------------------------------------------
1 | patient104/s0306lre
2 | patient105/s0303lre
3 | patient116/s0302lre
4 | patient117/s0291lre
5 | patient121/s0311lre
6 | patient122/s0312lre
7 | patient131/s0273lre
8 | patient150/s0287lre
9 | patient155/s0301lre
10 | patient156/s0299lre
11 | patient165/s0322lre
12 | patient166/s0275lre
13 | patient169/s0328lre
14 | patient170/s0274lre
15 | patient172/s0304lre
16 | patient173/s0305lre
17 | patient174/s0300lre
18 | patient180/s0374lre
19 | patient182/s0308lre
20 | patient184/s0363lre
21 | patient185/s0336lre
22 | patient198/s0402lre
23 | patient214/s0436_re
24 | patient229/s0452_re
25 | patient233/s0457_re
26 | patient234/s0460_re
27 | patient235/s0461_re
28 | patient236/s0462_re
29 | patient237/s0465_re
30 | patient238/s0466_re
31 | patient239/s0467_re
32 | patient240/s0468_re
33 | patient241/s0469_re
34 | patient242/s0471_re
35 | patient243/s0472_re
36 | patient244/s0473_re
37 | patient245/s0474_re
38 | patient246/s0478_re
39 | patient247/s0479_re
40 | patient248/s0481_re
41 | patient251/s0486_re
42 | patient252/s0487_re
43 | patient255/s0491_re
44 | patient260/s0496_re
45 | patient263/s0499_re
46 | patient264/s0500_re
47 | patient266/s0502_re
48 | patient267/s0504_re
49 | patient276/s0526_re
50 | patient277/s0527_re
51 | patient279/s0531_re
52 | patient284/s0543_re
53 |
54 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/r:
--------------------------------------------------------------------------------
1 |
2 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/sortann.m:
--------------------------------------------------------------------------------
1 | function varargout=sortann(varargin)
2 | %
3 | % sortann(recName,annName,beginTime,stopTime,outFile)
4 | %
5 | % Wrapper to WFDB SORTANN:
6 | % http://www.physionet.org/physiotools/wag/sortan-1.htm
7 | %
8 | % Rewrites the annotation file specified by recName and annName, arranging its contents in
9 | % canonical (time, num, and chan) order. The sorted (output) annotation file is always written
10 | % to the current directory. If the input annotation file is in the current directory, SORTANN
11 | % replaces it unless you specify a different output annotator name (using the outFile option).
12 | % If the input annotations are already in the correct order, no output is written unless you
13 | % have used the outFile option.
14 | %
15 | %
16 | %Input Parameters:
17 | % recName
18 | % String specifying the WFDB record file.
19 | %
20 | % annName
21 | % String specifying the reference WFDB annotation file.
22 | %
23 | % stopTime (Optional)
24 | % String specifying the start time in WFDB format (default is beginning of
25 | % record).
26 | %
27 | % stopTime (Optional)
28 | % String specifying the stop time in WFDB format (default is end of
29 | % record).
30 | %
31 | % outFile (Optional)
32 | % String specifying the output annotation file name.
33 | %
34 | %
35 | % Written by Ikaro Silva, 2013
36 | % Last Modified: -
37 | % Version 1.0
38 | % Since 0.9.0
39 | %
40 | % %Example (this will generate a /mitdb/100.sortedATR file at your directory):
41 | %
42 | % sortann('mitdb/100','atr',[],[],'sortedATR');
43 | % ann=rdann('mitdb/100','sortedATR');
44 | %
45 | %
46 | % See also RDANN
47 |
48 | %endOfHelp
49 | persistent javaWfdbExec
50 | if(isempty(javaWfdbExec))
51 | javaWfdbExec=getWfdbClass('sortann');
52 | end
53 |
54 | %Set default pararamter values
55 | inputs={'recName','annName','beginTime','stopTime','outFile'};
56 | recName=[];
57 | annName=[];
58 | beginTime=[];
59 | stopTime=[];
60 | outFile=[];
61 | for n=1:nargin
62 | if(~isempty(varargin{n}))
63 | eval([inputs{n} '=varargin{n};'])
64 | end
65 | end
66 |
67 | wfdb_argument={'-r',recName,'-a',annName};
68 |
69 | if(~isempty(beginTime))
70 | wfdb_argument{end+1}='-f';
71 | wfdb_argument{end+1}=beginTime;
72 | end
73 | if(~isempty(stopTime))
74 | wfdb_argument{end+1}='-t';
75 | wfdb_argument{end+1}=stopTime;
76 | end
77 | if(~isempty(outFile))
78 | wfdb_argument{end+1}='-o';
79 | wfdb_argument{end+1}=outFile;
80 | end
81 |
82 | javaWfdbExec.execToStringList(wfdb_argument);
83 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/sumann.m:
--------------------------------------------------------------------------------
1 | function varargout=sumann(varargin)
2 | %
3 | % report=sumann(recName,annName,stopTime,qrsAnnotationsOnly)
4 | %
5 | % Wrapper to WFDB SUMANN:
6 | % http://www.physionet.org/physiotools/wag/sumann-1.htm
7 | %
8 | % Reads a WFDB annotation file and summarize its contents.
9 | %
10 | % Ouput Parameters:
11 | %
12 | % report
13 | % String with the contaning summary of the contents, including the
14 | % number of annotations of each type as well the duration and number of
15 | % episodes of each rhythm and signal quality.
16 | %
17 | %Input Parameters:
18 | % recName
19 | % String specifying the WFDB record file.
20 | %
21 | % annName
22 | % String specifying the reference WFDB annotation file.
23 | %
24 | % stopTime (Optional)
25 | % String specifying the stop time in WFDB format (default is end of
26 | % record).
27 | %
28 | % qrsAnnotationsOnly (Optional)
29 | % 1x1 Boolean. If true, summarize QRS annotation only (default = 0).
30 | %
31 | %
32 | % Written by Ikaro Silva, 2013
33 | % Last Modified: -
34 | % Version 1.0
35 | % Since 0.9.0
36 | %
37 | % %Example (this will generate a /mitdb/100.qrs file at your directory):
38 | %
39 | % report=sumann('mitdb/100','atr');
40 | %
41 | %
42 | %
43 | % See also RDANN, MXM, WFDBTIME, BXB
44 |
45 | %endOfHelp
46 | persistent javaWfdbExec
47 | if(isempty(javaWfdbExec))
48 | javaWfdbExec=getWfdbClass('sumann');
49 | end
50 |
51 | %Set default pararamter values
52 | inputs={'recName','annName','stopTime','qrsAnnotationsOnly'};
53 | recName=[];
54 | annName=[];
55 | stopTime=[];
56 | qrsAnnotationsOnly=0;
57 | for n=1:nargin
58 | if(~isempty(varargin{n}))
59 | eval([inputs{n} '=varargin{n};'])
60 | end
61 | end
62 |
63 | wfdb_argument={'-r',recName,'-a',annName};
64 |
65 | if(~isempty(stopTime))
66 | wfdb_argument{end+1}='-t';
67 | wfdb_argument{end+1}=stopTime;
68 | end
69 | if(qrsAnnotationsOnly)
70 | wfdb_argument{end+1}='-q';
71 | end
72 |
73 | report=javaWfdbExec.execToStringList(wfdb_argument);
74 | if(nargout>0)
75 | varargout{1}=report;
76 | end
77 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/surrogate.m:
--------------------------------------------------------------------------------
1 | function Y=surrogate(x,M)
2 | %
3 | % Y=surrogate(x,M)
4 | %
5 | % Generates M amplitude adjusted phase shuffled surrogate time series from x.
6 | % Useufel for testing the underlying assumption that the null hypothesis consists
7 | % of linear dynamics with possibly non-linear, monotonically increasing,
8 | % measurement function.
9 | %
10 | % Required Input Parameters:
11 | %
12 | % x
13 | % Nx1 vector of doubles
14 | %
15 | % M
16 | % 1x1 scalar specifying the number of surrogate time series to
17 | % generate.
18 | %
19 | % Required Output Parameters:
20 | %
21 | % Y
22 | % NxM vector of doubles
23 | %
24 | %
25 | %
26 | % References:
27 | %
28 | %[1] Kaplan, Daniel, and Leon Glass. Understanding nonlinear dynamics. Vol. 19. Springer, 1995.
29 | %
30 | %
31 | % Written by Ikaro Silva, 2014
32 | % Last Modified: November 20, 2014
33 | % Version 1.0
34 | % Since 0.9.8
35 | %
36 | %
37 | %
38 | %
39 | % See also MSENTROPY, SURROGATE
40 |
41 | %endOfHelp
42 |
43 | % 1. Amp transform original data to Gaussian distribution
44 | % 2. Phase randomize #1
45 | % 3. Amp transform #2 to original
46 | % Auto-correlation function should be similar but not exact!
47 |
48 | x=x(:);
49 | N=length(x);
50 | Y=zeros(N,M);
51 |
52 | for m=1:M
53 |
54 | %Step 1
55 | y=randn(N,1);
56 | y=amplitudeTransform(x,y,N);
57 |
58 | %Step 2
59 | y=phaseShuffle(y,N);
60 |
61 | %Step 3
62 | y=amplitudeTransform(y,x,N);
63 | Y(:,m)=y;
64 | end
65 |
66 |
67 | %%% Helper functions
68 |
69 | function target=amplitudeTransform(x,target,N)
70 |
71 | %Steps:
72 | %1. Sort the source by increasing amp
73 | %2. Sort target as #1
74 | %3. Swap source amp by target amp
75 | %4. Sort #3 by increasing time index of #1
76 | X=[[1:N]' x];
77 | X=sortrows(X,2);
78 | target=[X(:,1) sort(target)];
79 | target=sortrows(target,1);
80 | target=target(:,2);
81 |
82 |
83 |
84 | function y=phaseShuffle(x,N)
85 |
86 | %%Shuffle spectrum
87 | X=fft(x);
88 | Y=X;
89 | mid=floor(N/2)+ mod(N,2);
90 | phi=2*pi*rand(mid-1,1); %Generate random phase
91 | Y(2:mid)=abs(X(2:mid)).*cos(phi) + j*abs(X(2:mid)).*sin(phi);
92 | if(~mod(N,2))
93 | %Even series has Nyquist in the middle+1 because of DC
94 | Y(mid+2:end)=conj(flipud(Y(2:mid)));
95 | Y(mid+1)=X(mid+1);
96 | else
97 | %Odd series is fully symetric except for DC
98 | Y(mid+1:end)=conj(flipud(Y(2:mid)));
99 | end
100 |
101 | y=real(ifft(Y));
102 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/visgraph.m:
--------------------------------------------------------------------------------
1 | function varargout=visgraph(varargin)
2 | %
3 | % [k,logP]=visgraph(x)
4 | %
5 | % Visibility Graph analysis of a time series as described in:
6 | %
7 | % Lacasa, Lucas, et al.
8 | % "From time series to complex networks: The visibility graph."
9 | % Proceedings of the National Academy of Sciences 105.13 (2008): 4972-4975.
10 | %
11 | % Required input parameter:
12 | % x
13 | % Nx1 matrix (doubles) of time series to be analyzed.
14 | %
15 | %
16 | %
17 | % Written by Ikaro Silva, 20134
18 | % Last Modified: November 24, 2014
19 | % Version 1.0
20 | %
21 | % Since 0.9.8
22 | %
23 | %
24 | % %Example
25 | % %Generate Conway Series
26 | % N=1000;
27 | % a=ones(N,1);
28 | % out=ones(N,1);
29 | % for n=3:N
30 | % a(n)=a(a(n-1))+ a(n-a(n-1));
31 | % out(n)= a(n) - (n/2);
32 | % end
33 | %
34 | % %Generate Surrogate Data
35 | % nS=5;
36 | % S=surrogate(out,nS);
37 | % subplot(3,1,1)
38 | % plot(out);title('Conway Series')
39 | % subplot(3,1,2)
40 | % plot(S(:,1),'r');title('Amplitude Adjusted Surrogate Data')
41 | %
42 | % %Calculate visibility graph for all series
43 | % [k,logP]=visgraph(out);
44 | % subplot(3,1,3)
45 | % plot(k,logP);hold on;grid on
46 | %
47 | % for n=1:nS
48 | % [k,logP]=visgraph(S(:,n));
49 | % subplot(3,1,3)
50 | % plot(k,logP,'r');
51 | % end
52 | %
53 | % See also SURROGATE, DFA, MSENTROPY, CORRINT
54 |
55 | %endOfHelp
56 |
57 | persistent javaWfdbExec config
58 | if(isempty(javaWfdbExec))
59 | [javaWfdbExec,config]=getWfdbClass('visbility');
60 | end
61 |
62 | %Set default pararamter values
63 | inputs={'x'};
64 | outputs={'k','logP'};
65 | k=[];
66 | logP=[];
67 | for n=1:nargin
68 | if(~isempty(varargin{n}))
69 | eval([inputs{n} '=varargin{n};'])
70 | end
71 | end
72 |
73 | if(config.inOctave)
74 | x=cellstr(num2str(x));
75 | x=java2mat(javaWfdbExec.execWithStandardInput(x));
76 | Nx=x.size;
77 | out=cell(Nx,1);
78 | for n=1:Nx
79 | out{n}=x.get(n-1);
80 | end
81 | else
82 | out=cell(javaWfdbExec.execWithStandardInput(x).toArray);
83 | end
84 |
85 | M=length(out);
86 | k=zeros(M,1)+NaN;
87 | logP=zeros(M,1)+NaN;
88 | if(length(out{end})==1)
89 | out(end)=[];
90 | M=M-1;
91 | end
92 | for m=1:M
93 | str=out{m};
94 | sep=regexp(str,'\s');
95 | k(m)=str2num(str(1:sep));
96 | logP(m)=str2num(str(sep(1):end));
97 | end
98 |
99 | for n=1:nargout
100 | eval(['varargout{n}=' outputs{n} ';'])
101 | end
102 |
103 |
104 |
105 |
106 |
107 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/wfdb-app-JVM7-0-10-0.jar:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/wfdb-app-JVM7-0-10-0.jar
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/wfdb.m:
--------------------------------------------------------------------------------
1 | function wfdb
2 | % wfdb
3 | %
4 | %Display list of all function available for the WFDB App Toolbox.
5 | %
6 | %Since 0.0.1
7 | %
8 | %%Example:
9 | % wfdb
10 | %
11 | [~,config]=wfdbloadlib;
12 | help(config.MATLAB_PATH(1:end-1))
13 | %Display information regarding the WFDB Toolbox.
14 | %Written by Ikaro Silva 2012
15 |
16 |
17 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/wfdb2mat.m:
--------------------------------------------------------------------------------
1 | function wfdb2mat(varargin)
2 | %
3 | % wfdm2mat(recordName,signaList,N,N0)
4 | %
5 | % Wrapper to WFDB WFDB2MAT:
6 | % http://physionet.org/physiotools/wag/wfdb2m-1.htm
7 | %
8 | % Converts a WFDB-compatible signal file to MATLAB/Octave *.mat file.
9 | % The output files are recordNamem.mat and recordNamem.hea. The standard output of
10 | % WFDB2MAT will be saved in a file recordNamem.info.
11 | %
12 | % Required Parameters:
13 | %
14 | % recorName
15 | % String specifying the name of the record in the WFDB path or
16 | % in the current directory.
17 | %
18 | % Optional Parameters are:
19 | %
20 | % signalList
21 | % A Mx1 array of integers. Read only the signals (columns)
22 | % named in the signalList (default: read all signals).
23 | % N
24 | % A 1x1 integer specifying the sample number at which to stop reading the
25 | % record file (default read all the samples = N).
26 | % N0
27 | % A 1x1 integer specifying the sample number at which to start reading the
28 | % record file (default 1 = first sample).
29 | %
30 | %
31 | %
32 | % NOTE:
33 | % You can use the WFDB2MAT command in order to convert the record data into a *.mat file,
34 | % which can then be loaded into MATLAB/Octave's workspace using the LOAD command.
35 | % This will load the signal data in raw units (use RDMAT to load the signal in physical units).
36 | %
37 | %
38 | % Written by Ikaro Silva, 2014
39 | % Last Modified: September 15, 2014
40 | % Version 0.1
41 | %
42 | % Since 0.9.7
43 | %
44 | % %Example:
45 | % wfdb2mat('mitdb/200')
46 | % [tm,signal,Fs,labels]=rdmat('200m');
47 | %
48 | %
49 | % See also RDSAMP, RDMAT
50 |
51 | %endOfHelp
52 |
53 | persistent javaWfdbExec config
54 | if(isempty(javaWfdbExec))
55 | [javaWfdbExec,config]=getWfdbClass('wfdb2mat');
56 | end
57 |
58 | %Set default pararameter values
59 | inputs={'recordName','signalList','N','N0'};
60 | signalList=[];
61 | N=[];
62 | N0=1;
63 | for n=1:nargin
64 | if(~isempty(varargin{n}))
65 | eval([inputs{n} '=varargin{n};'])
66 | end
67 | end
68 |
69 | wfdb_argument={'-r',recordName,'-f',['s' num2str(N0-1)]};
70 |
71 | if(~isempty(N))
72 | wfdb_argument{end+1}='-t';
73 | wfdb_argument{end+1}=['s' num2str(N)];
74 | end
75 |
76 | if(~isempty(signalList))
77 | wfdb_argument{end+1}='-s ';
78 | %-1 is necessary because WFDB is 0 based indexed.
79 | for sInd=1:length(signalList)
80 | wfdb_argument{end+1}=[num2str(signalList(sInd)-1)];
81 | end
82 | end
83 |
84 | data=javaWfdbExec.execToStringList(wfdb_argument);
85 |
86 |
87 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/wfdbRecordViewer.fig:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/wfdbRecordViewer.fig
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Biyomedikal_-saretler_Kullanilarak_Yapay_Zeka_ile_Hastalik_Teshisi/mcode/wfdbdemo.m:
--------------------------------------------------------------------------------
1 | function wfdbdemo()
2 | % WFDB App Toolbox Demo
3 | %
4 | % Written by Ikaro Silva, 2013
5 | % Last modified: January 10, 2014
6 | %
7 |
8 | [~,config]=wfdbloadlib;
9 | echo on
10 | display('Reading samples ECG signal from MIT-BIH Arrhythmia Database')
11 | N=10000;
12 | [ecg,Fs,tm]=rdsamp('mitdb/100',1,N);
13 |
14 | display('Reading and plotting annotations (human labels) of QRS complexes performend on the signals')
15 | %by cardiologists.
16 | [ann,type,subtype,chan,num]=rdann('mitdb/100','atr',1,N);
17 |
18 | %Plot 2D version of signal and labels
19 | figure
20 | plot(tm(1:N),ecg(1:N));hold on;grid on
21 | plot(tm(ann(ann RecycleNet kaynak kodları yakında bu repoya eklenecektir!
24 |
25 |
26 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Derin_Ogrenme_ile_Geri_Donusum_Malzemelerinin_Taninmasi/RecycleNet-DeepCon18.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Derin_Ogrenme_ile_Geri_Donusum_Malzemelerinin_Taninmasi/RecycleNet-DeepCon18.pdf
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Derin_Ogrenme_ile_Geri_Donusum_Malzemelerinin_Taninmasi/RecycleNet.PNG:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Derin_Ogrenme_ile_Geri_Donusum_Malzemelerinin_Taninmasi/RecycleNet.PNG
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Derin_Pekistirmeli_Ogrenmeye_Giris_Doom_Oynayan_Ajan_Gelistirme/.gitignore:
--------------------------------------------------------------------------------
1 | # Byte-compiled / optimized / DLL files
2 | __pycache__/
3 | *.py[cod]
4 | *$py.class
5 |
6 | # C extensions
7 | *.so
8 |
9 | # Distribution / packaging
10 | .Python
11 | build/
12 | develop-eggs/
13 | dist/
14 | downloads/
15 | eggs/
16 | .eggs/
17 | lib/
18 | lib64/
19 | parts/
20 | sdist/
21 | var/
22 | wheels/
23 | *.egg-info/
24 | .installed.cfg
25 | *.egg
26 | MANIFEST
27 |
28 | # PyInstaller
29 | # Usually these files are written by a python script from a template
30 | # before PyInstaller builds the exe, so as to inject date/other infos into it.
31 | *.manifest
32 | *.spec
33 |
34 | # Installer logs
35 | pip-log.txt
36 | pip-delete-this-directory.txt
37 |
38 | # Unit test / coverage reports
39 | htmlcov/
40 | .tox/
41 | .coverage
42 | .coverage.*
43 | .cache
44 | nosetests.xml
45 | coverage.xml
46 | *.cover
47 | .hypothesis/
48 | .pytest_cache/
49 |
50 | # Translations
51 | *.mo
52 | *.pot
53 |
54 | # Django stuff:
55 | *.log
56 | local_settings.py
57 | db.sqlite3
58 |
59 | # Flask stuff:
60 | instance/
61 | .webassets-cache
62 |
63 | # Scrapy stuff:
64 | .scrapy
65 |
66 | # Sphinx documentation
67 | docs/_build/
68 |
69 | # PyBuilder
70 | target/
71 |
72 | # Jupyter Notebook
73 | .ipynb_checkpoints
74 |
75 | # pyenv
76 | .python-version
77 |
78 | # celery beat schedule file
79 | celerybeat-schedule
80 |
81 | # SageMath parsed files
82 | *.sage.py
83 |
84 | # Environments
85 | .env
86 | .venv
87 | env/
88 | venv/
89 | ENV/
90 | env.bak/
91 | venv.bak/
92 |
93 | # Spyder project settings
94 | .spyderproject
95 | .spyproject
96 |
97 | # Rope project settings
98 | .ropeproject
99 |
100 | # mkdocs documentation
101 | /site
102 |
103 | # mypy
104 | .mypy_cache/
105 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Derin_Pekistirmeli_Ogrenmeye_Giris_Doom_Oynayan_Ajan_Gelistirme/README.md:
--------------------------------------------------------------------------------
1 | # DeepCon18 Deep RL Atolyesi
2 | 05-06 Ekim tarihlerinde gerçekleşen DeepCon konferansındaki Derin Pekiştirmeli Öğrenme Atölyesi
3 | Kaynak: https://github.com/FurkanArslan/deepcon-deep-rl-workshop
4 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Derin_Pekistirmeli_Ogrenmeye_Giris_Doom_Oynayan_Ajan_Gelistirme/_config.yml:
--------------------------------------------------------------------------------
1 | theme: jekyll-theme-cayman
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Derin_Pekistirmeli_Ogrenmeye_Giris_Doom_Oynayan_Ajan_Gelistirme/ayarlar/basic.cfg:
--------------------------------------------------------------------------------
1 | # Lines starting with # are treated as comments (or with whitespaces+#).
2 | # It doesn't matter if you use capital letters or not.
3 | # It doesn't matter if you use underscore or camel notation for keys, e.g. episode_timeout is the same as episodeTimeout.
4 |
5 | doom_scenario_path = basic.wad
6 | doom_map = map01
7 |
8 | # Rewards
9 | living_reward = -1
10 |
11 | # Rendering options
12 | screen_resolution = RES_320X240
13 | screen_format = GRAY8
14 | render_hud = True
15 | render_crosshair = false
16 | render_weapon = true
17 | render_decals = false
18 | render_particles = false
19 | window_visible = false
20 |
21 | # make episodes start after 20 tics (after unholstering the gun)
22 | episode_start_time = 14
23 |
24 | # make episodes finish after 300 actions (tics)
25 | episode_timeout = 300
26 |
27 | # Available buttons
28 | available_buttons =
29 | {
30 | MOVE_LEFT
31 | MOVE_RIGHT
32 | ATTACK
33 | }
34 |
35 | # Game variables that will be in the state
36 | available_game_variables = { AMMO2}
37 |
38 | mode = PLAYER
39 | doom_skill = 5
40 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Derin_Pekistirmeli_Ogrenmeye_Giris_Doom_Oynayan_Ajan_Gelistirme/models/checkpoint:
--------------------------------------------------------------------------------
1 | model_checkpoint_path: "model.ckpt"
2 | all_model_checkpoint_paths: "model.ckpt"
3 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Derin_Pekistirmeli_Ogrenmeye_Giris_Doom_Oynayan_Ajan_Gelistirme/models/model.ckpt.data-00000-of-00001:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Derin_Pekistirmeli_Ogrenmeye_Giris_Doom_Oynayan_Ajan_Gelistirme/models/model.ckpt.data-00000-of-00001
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Derin_Pekistirmeli_Ogrenmeye_Giris_Doom_Oynayan_Ajan_Gelistirme/models/model.ckpt.index:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Derin_Pekistirmeli_Ogrenmeye_Giris_Doom_Oynayan_Ajan_Gelistirme/models/model.ckpt.index
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Derin_Pekistirmeli_Ogrenmeye_Giris_Doom_Oynayan_Ajan_Gelistirme/models/model.ckpt.meta:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Derin_Pekistirmeli_Ogrenmeye_Giris_Doom_Oynayan_Ajan_Gelistirme/models/model.ckpt.meta
--------------------------------------------------------------------------------
/Atolye_Dosyalari/E-Ticaret_ve_Yapay_Zeka_Uygulamalari/E-Ticaret ve Yapay Zeka Uygulamaları_CüneytAksakallı.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/E-Ticaret_ve_Yapay_Zeka_Uygulamalari/E-Ticaret ve Yapay Zeka Uygulamaları_CüneytAksakallı.pdf
--------------------------------------------------------------------------------
/Atolye_Dosyalari/E-Ticaret_ve_Yapay_Zeka_Uygulamalari/README.md:
--------------------------------------------------------------------------------
1 | # DeepCon18 / E-Ticaret ve Yapay Zeka Uygulamalari
2 | http://derindelimavi.blogspot.com/2018/10/deepcon18-payment-anomaly-workshop.html
3 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/E-Ticaret_ve_Yapay_Zeka_Uygulamalari/best_comp.h5:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/E-Ticaret_ve_Yapay_Zeka_Uygulamalari/best_comp.h5
--------------------------------------------------------------------------------
/Atolye_Dosyalari/E-Ticaret_ve_Yapay_Zeka_Uygulamalari/circular_normalization.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/E-Ticaret_ve_Yapay_Zeka_Uygulamalari/circular_normalization.png
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Göktuğ İslamoğlu DeepCon 18.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Göktuğ İslamoğlu DeepCon 18.pdf
--------------------------------------------------------------------------------
/Atolye_Dosyalari/HalilİbrahimÇelenli_DeepCon.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/HalilİbrahimÇelenli_DeepCon.pdf
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Mamografi_Goruntuleri_Kullanarak_Meme_Kanseri_Teshisinin_Performans_Degerlendirmesi/ReadMe.md:
--------------------------------------------------------------------------------
1 |
2 | ## Mamografi Görüntüleri Kullanarak Meme Kanseri Teşhisinin Performans Değerlendirmesi
3 | ### Burcu Bektaş
4 | DeepCon'18 Atölye Çalışması
5 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Mamografi_Goruntuleri_Kullanarak_Meme_Kanseri_Teshisinin_Performans_Degerlendirmesi/lablenumeric.xlsx:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Mamografi_Goruntuleri_Kullanarak_Meme_Kanseri_Teshisinin_Performans_Degerlendirmesi/lablenumeric.xlsx
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Mamografi_Goruntuleri_Kullanarak_Meme_Kanseri_Teshisinin_Performans_Degerlendirmesi/mat2arff.m:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Mamografi_Goruntuleri_Kullanarak_Meme_Kanseri_Teshisinin_Performans_Degerlendirmesi/mat2arff.m
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Mamografi_Goruntuleri_Kullanarak_Meme_Kanseri_Teshisinin_Performans_Degerlendirmesi/onisleme.m:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Mamografi_Goruntuleri_Kullanarak_Meme_Kanseri_Teshisinin_Performans_Degerlendirmesi/onisleme.m
--------------------------------------------------------------------------------
/Atolye_Dosyalari/OpenAI_Gym_ile_Pekistirmeli_Ogrenme/README.md:
--------------------------------------------------------------------------------
1 | # DeepCon'18 Deep Reinforcement Learning Workshop
2 |
3 |
4 | ## Katılım için gereksinimler
5 |
6 | Workshop katılımı için temel Python bilgisi gerekmektedir. İnteraktif bir çalışma olacağı için laptopunuzla beraber gelmelisiniz, en geç 30 eylül tarihine kadar hazır olacak bu repoda DQN ajanının kullanım için hazır bir sınıfını bulacaksınız. Bu sınıfı openai gym cartpole ortamında nasıl kullanacağımızı ve bu sınıfın detaylarını birlikte inceleyeceğiz . [Python](https://www.python.org/downloads/) adresinden indirebileceğiniz Python'un son versiyonu bilgisayarınızda kurulu olmalıdır. Editör olarak [Pycharm](https://www.jetbrains.com/pycharm/) kullanacağım. Bu aşamalar tamamlandıktan sonra aşağıdaki kodu bilgisayarınızda çalıştırmanız zaman kazanmamız açısından önemlidir.
7 | ```
8 | git clone https://github.com/umutcanaltin/deepcon-drl-workshop.git
9 | pip install numpy matplotlib gym keras
10 | ```
11 |
12 | Görüşmek üzere!
13 | Umut Can ALTIN
14 | https://umutcanaltin.github.io/
15 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/OpenAI_Gym_ile_Pekistirmeli_Ogrenme/cartpole_örnek_hareketler.py:
--------------------------------------------------------------------------------
1 | import gym
2 | env = gym.make('CartPole-v0')
3 |
4 | env.reset()
5 | rewards = []
6 |
7 | episodes=10
8 | max_time=100
9 | for e in range(episodes):
10 | env.render()
11 | for t in range(max_time) :
12 | state, reward, done, info = env.step(env.action_space.sample()) # rastgele hareket et
13 |
14 | if done:
15 | rewards.append(t)
16 | env.reset()
17 |
18 |
19 |
20 | print(rewards)
21 | env.close()
--------------------------------------------------------------------------------
/Atolye_Dosyalari/OpenAI_Gym_ile_Pekistirmeli_Ogrenme/model.h5:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/OpenAI_Gym_ile_Pekistirmeli_Ogrenme/model.h5
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Pekistirmeli_Ogrenmede_Cok_Kollu_Haydutlar_Problemi/.gitignore:
--------------------------------------------------------------------------------
1 | # Byte-compiled / optimized / DLL files
2 | __pycache__/
3 | *.py[cod]
4 | *$py.class
5 |
6 | # C extensions
7 | *.so
8 |
9 | # Distribution / packaging
10 | .Python
11 | build/
12 | develop-eggs/
13 | dist/
14 | downloads/
15 | eggs/
16 | .eggs/
17 | lib/
18 | lib64/
19 | parts/
20 | sdist/
21 | var/
22 | wheels/
23 | *.egg-info/
24 | .installed.cfg
25 | *.egg
26 | MANIFEST
27 |
28 | # PyInstaller
29 | # Usually these files are written by a python script from a template
30 | # before PyInstaller builds the exe, so as to inject date/other infos into it.
31 | *.manifest
32 | *.spec
33 |
34 | # Installer logs
35 | pip-log.txt
36 | pip-delete-this-directory.txt
37 |
38 | # Unit test / coverage reports
39 | htmlcov/
40 | .tox/
41 | .nox/
42 | .coverage
43 | .coverage.*
44 | .cache
45 | nosetests.xml
46 | coverage.xml
47 | *.cover
48 | .hypothesis/
49 | .pytest_cache/
50 |
51 | # Translations
52 | *.mo
53 | *.pot
54 |
55 | # Django stuff:
56 | *.log
57 | local_settings.py
58 | db.sqlite3
59 |
60 | # Flask stuff:
61 | instance/
62 | .webassets-cache
63 |
64 | # Scrapy stuff:
65 | .scrapy
66 |
67 | # Sphinx documentation
68 | docs/_build/
69 |
70 | # PyBuilder
71 | target/
72 |
73 | # Jupyter Notebook
74 | .ipynb_checkpoints
75 |
76 | # IPython
77 | profile_default/
78 | ipython_config.py
79 |
80 | # pyenv
81 | .python-version
82 |
83 | # celery beat schedule file
84 | celerybeat-schedule
85 |
86 | # SageMath parsed files
87 | *.sage.py
88 |
89 | # Environments
90 | .env
91 | .venv
92 | env/
93 | venv/
94 | ENV/
95 | env.bak/
96 | venv.bak/
97 |
98 | # Spyder project settings
99 | .spyderproject
100 | .spyproject
101 |
102 | # Rope project settings
103 | .ropeproject
104 |
105 | # mkdocs documentation
106 | /site
107 |
108 | # mypy
109 | .mypy_cache/
110 | .dmypy.json
111 | dmypy.json
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Pekistirmeli_Ogrenmede_Cok_Kollu_Haydutlar_Problemi/Pekiştirmeli Öğrenmede Çok Kollu Haydutlar Problemi.pptx:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Pekistirmeli_Ogrenmede_Cok_Kollu_Haydutlar_Problemi/Pekiştirmeli Öğrenmede Çok Kollu Haydutlar Problemi.pptx
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Pekistirmeli_Ogrenmede_Cok_Kollu_Haydutlar_Problemi/README.md:
--------------------------------------------------------------------------------
1 | # DeepCon - Pekiştirmeli Öğrenmede Çok Kollu Haydutlar Problemi
2 |
3 | ## Katılım için gereksinimler
4 |
5 | Workshop katılımı için temel Python bilgisi gerekmektedir. İnteraktif bir çalışma olacağı için laptopunuzla beraber gelmelisiniz. [Python](https://www.python.org/downloads/) adresinden indirebileceğiniz Python'un son versiyonu bilgisayarınızda kurulu olmalıdır. Bu aşamalar tamamlandıktan sonra aşağıdaki kodu bilgisayarınızda çalıştırmanız zaman kazanmamız açısından önemlidir. Kod yazımı kolaylığı için [Visual Studio Code](https://code.visualstudio.com/download) kullanmanızı öneriyoruz. Bu repository'yi alabilmek için [Git](https://git-scm.com/downloads) indirmeniz gerekmektedir.
6 |
7 | ```
8 | git clone https://github.com/kivancguckiran/deepcon-rl-workshop
9 | pip install numpy matplotlib
10 | ```
11 |
12 | Görüşmek üzere!
13 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Pekistirmeli_Ogrenmede_Cok_Kollu_Haydutlar_Problemi/min_util.py:
--------------------------------------------------------------------------------
1 | import numpy as np
2 | import matplotlib.pyplot as plt
3 |
4 |
5 | class bandit:
6 | def __init__(self, eps=0, step_size=0., initial=0, variance=1):
7 | self.epsilon = eps
8 | self.step_size = step_size
9 | self.k_arm = 5
10 | self.variance = variance
11 |
12 | self.q_true = np.asarray([-0.25, 2, 1, -1.5, 3])
13 | self.q_estimates = np.zeros(self.k_arm) + initial
14 |
15 | self.actions_taken = np.zeros(self.k_arm)
16 |
17 | def action(self):
18 | if np.random.rand() < self.epsilon:
19 | return np.random.randint(self.k_arm)
20 | else:
21 | return np.argmax(self.q_estimates)
22 |
23 | def step(self):
24 | idx = self.action()
25 |
26 | self.actions_taken[idx] += 1
27 |
28 | reward = self.q_true[idx] + np.random.randn() * self.variance
29 |
30 | if self.step_size == 0:
31 | self.q_estimates[idx] = self.q_estimates[idx] + (1.0 / self.actions_taken[idx]) * (reward - self.q_estimates[idx])
32 | else:
33 | self.q_estimates[idx] = self.q_estimates[idx] + self.step_size * (reward - self.q_estimates[idx])
34 |
35 | return reward
36 |
37 | def change_q_true(self):
38 | self.q_true = np.asarray([1, -0.5, -2, 2, 0.25])
39 |
40 | def take_steps(self, count):
41 | for _ in np.arange(count):
42 | self.step()
43 |
44 | self.showdown()
45 |
46 | def plot(self):
47 | plt.violinplot(positions=np.arange(self.k_arm), dataset=self.q_true + np.random.randn(100,self.k_arm) * self.variance)
48 | plt.plot(self.q_estimates, color='red', marker='o', linestyle='', markersize=5)
49 | plt.show()
50 |
51 | def dump(self):
52 | for i in np.arange(self.k_arm):
53 | print('Arm #%u: %u times.' % (i, self.actions_taken[i]))
54 |
55 | print('Best Arm: #%u.' % (np.argmax(self.q_true)))
56 |
57 | def showdown(self):
58 | self.dump()
59 | self.plot()
60 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Pekistirmeli_Ogrenmede_Cok_Kollu_Haydutlar_Problemi/r.md:
--------------------------------------------------------------------------------
1 |
2 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Pekistirmeli_Ogrenmede_Cok_Kollu_Haydutlar_Problemi/util.py:
--------------------------------------------------------------------------------
1 | import numpy as np
2 | import matplotlib.pyplot as plt
3 |
4 |
5 | class bandit:
6 | def __init__(self, k_arm=10, eps=0, step_size=0.1):
7 | self.k_arm = k_arm
8 | self.epsilon = eps
9 | self.step_size = step_size
10 |
11 | self.q_true = np.random.randn(self.k_arm)
12 | self.q_estimates = np.zeros(self.k_arm)
13 |
14 | self.best_arm = np.argmax(self.q_true)
15 |
16 | def action(self):
17 | if np.random.rand() < self.epsilon:
18 | return np.random.randint(self.k_arm)
19 |
20 | return np.argmax(self.q_estimates)
21 |
22 | def step(self):
23 | idx = self.action()
24 |
25 | reward = self.q_true[idx] + np.random.randn()
26 |
27 | self.q_estimates[idx] += self.step_size * (reward - self.q_estimates[idx])
28 |
29 | return reward
30 |
31 | def take_steps(self, count):
32 | for _ in np.arange(count):
33 | self.step()
34 |
35 | self.plot_current_estimates()
36 |
37 | def plot_current_estimates(self):
38 | plt.violinplot(positions=np.arange(self.k_arm), dataset=self.q_true + np.random.randn(100,10))
39 | plt.plot(self.q_estimates, color='red', marker='o', linestyle='', markersize=5)
40 | plt.show()
41 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/ReadMe.md:
--------------------------------------------------------------------------------
1 | ### DeepCon'18 Yapay Zeka Konferansı
2 | 📚💻Atölye çalışma dosyalarına [buradan](https://github.com/deeplearningturkiye/DeepCon18/blob/master/README.md) ulaşabilirsiniz. 🎯
3 |
4 |
5 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/README.md:
--------------------------------------------------------------------------------
1 | # DeepCon'18 - Atölye - TensorFlow.js
2 | [Yapay Zeka Konferansı (DeepCon'18)](http://deepcon.deeplearningturkiye.com/) **"TensorFlow.js ile Derin Öğrenme Web Uygulaması Geliştirme"** Atölye Sunum ve Demo Kaynak Kodları
3 |
4 | [](http://deepcon.deeplearningturkiye.com/)
5 |
6 | ## Online Demo
7 | Atölye uygulamasını online olarak test etmek için [tıklayınız](https://yavuzkomecoglu.github.io/deepcon_demo/).
8 |
9 | ## Sunum
10 | - PowerPoint versiyonu için [tıklayınız.](https://github.com/yavuzKomecoglu/deepcon18_atolye_tensorflowJS/blob/master/DeepCon18_sunum.ppt)
11 | - PDF versiyonu için [tıklayınız.](https://github.com/yavuzKomecoglu/deepcon18_atolye_tensorflowJS/blob/master/DeepCon18_sunum.pdf)
12 |
13 | ## Faydalı Kaynaklar
14 | ##### Resmi Kaynaklar
15 |
16 | - [Resmi TensorFlow.js Sitesi](https://js.tensorflow.org/)
17 | - [TensorFlow.js Örnekleri](https://github.com/tensorflow/tfjs-examples)
18 | - [TensorFlow.js Giriş](https://medium.com/tensorflow/introducing-tensorflow-js-machine-learning-in-javascript-bf3eab376db)
19 | - [Yapay Sinir Ağları Görselleştirme Demosu](https://playground.tensorflow.org)
20 |
21 | ##### Diğer Kaynaklar
22 | - [Cheetsheet](https://towardsdatascience.com/50-tensorflow-js-api-explained-in-5-minutes-tensorflow-js-cheetsheet-4f8c7f9cc8b2)
23 | - [Deep Learning Türkiye Blog - TensorFlow.js ile Postür Tahmini için Derin Öğrenme – Sıddık Açıl](https://medium.com/deep-learning-turkiye/tensorflow-js-ile-post%C3%BCr-tahmini-i%C3%A7in-derin-%C3%B6%C4%9Frenme-l-f040de7355d8)
24 |
25 | ### İletişim
26 | Sosyal medya hesaplarım ve bloguma [yavuzkomecoglu.com](http://yavuzkomecoglu.com/) üzerinden ulaşabilirsiniz.
27 | Her türlü soru ve iletişim için: [komecoglu.yavuz@gmail.com](mailto:komecoglu.yavuz@gmail.com)
28 |
29 |
30 |
31 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/YavuzKomecoglu_DeepCon18_Sunum.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/YavuzKomecoglu_DeepCon18_Sunum.pdf
--------------------------------------------------------------------------------
/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/YavuzKomecoglu_DeepCon18_Sunum.ppt:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/YavuzKomecoglu_DeepCon18_Sunum.ppt
--------------------------------------------------------------------------------
/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/images/angry.gif:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/images/angry.gif
--------------------------------------------------------------------------------
/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/images/excited.gif:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/images/excited.gif
--------------------------------------------------------------------------------
/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/images/laugh.gif:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/images/laugh.gif
--------------------------------------------------------------------------------
/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/images/out.gif:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/images/out.gif
--------------------------------------------------------------------------------
/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/images/sad.gif:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/images/sad.gif
--------------------------------------------------------------------------------
/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/images/sleep.gif:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/images/sleep.gif
--------------------------------------------------------------------------------
/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/sounds/shutter.mp3:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/sounds/shutter.mp3
--------------------------------------------------------------------------------
/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/sounds/shutter.ogg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/sounds/shutter.ogg
--------------------------------------------------------------------------------
/DEEPCON18.jpeg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/DEEPCON18.jpeg
--------------------------------------------------------------------------------
/Konusmaci_Sunumlari/1-DeepCon_AykutErdem.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Konusmaci_Sunumlari/1-DeepCon_AykutErdem.pdf
--------------------------------------------------------------------------------
/Konusmaci_Sunumlari/2-DeepCon_Onur Koc.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Konusmaci_Sunumlari/2-DeepCon_Onur Koc.pdf
--------------------------------------------------------------------------------
/Konusmaci_Sunumlari/3-DeepCon_Av.SertalSiraci.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Konusmaci_Sunumlari/3-DeepCon_Av.SertalSiraci.pdf
--------------------------------------------------------------------------------
/Konusmaci_Sunumlari/4-DeepCon_AliOsmanOrs.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Konusmaci_Sunumlari/4-DeepCon_AliOsmanOrs.pdf
--------------------------------------------------------------------------------
/Konusmaci_Sunumlari/5-DeepCon_ESerdarGokpinar.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Konusmaci_Sunumlari/5-DeepCon_ESerdarGokpinar.pdf
--------------------------------------------------------------------------------
/Konusmaci_Sunumlari/6-DeepCon_BerkSunar.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Konusmaci_Sunumlari/6-DeepCon_BerkSunar.pdf
--------------------------------------------------------------------------------
/Konusmaci_Sunumlari/7-DeepCon_AlperGerçek.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Konusmaci_Sunumlari/7-DeepCon_AlperGerçek.pdf
--------------------------------------------------------------------------------
/Konusmaci_Sunumlari/9-DeepCon_ErdemYoruk.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Konusmaci_Sunumlari/9-DeepCon_ErdemYoruk.pdf
--------------------------------------------------------------------------------
/Konusmaci_Sunumlari/ReadMe.md:
--------------------------------------------------------------------------------
1 |
2 | ## DeepCon'18 Yapay Zeka Konferansı
3 | ### 5 Ekim 2018 tarihinde yapılan konuşmaların sunum dosyalarına bu dosyadan ulaşabilirsiniz.
4 |
--------------------------------------------------------------------------------
/atolye_egitmenleri1.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/atolye_egitmenleri1.png
--------------------------------------------------------------------------------
/atolye_egitmenleri2.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/atolye_egitmenleri2.png
--------------------------------------------------------------------------------
/atolye_egitmenleri3.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/atolye_egitmenleri3.png
--------------------------------------------------------------------------------
/deepcon_logo.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/deepcon_logo.png
--------------------------------------------------------------------------------
26 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Derin_Ogrenme_ile_Geri_Donusum_Malzemelerinin_Taninmasi/RecycleNet-DeepCon18.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Derin_Ogrenme_ile_Geri_Donusum_Malzemelerinin_Taninmasi/RecycleNet-DeepCon18.pdf
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Derin_Ogrenme_ile_Geri_Donusum_Malzemelerinin_Taninmasi/RecycleNet.PNG:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Derin_Ogrenme_ile_Geri_Donusum_Malzemelerinin_Taninmasi/RecycleNet.PNG
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Derin_Pekistirmeli_Ogrenmeye_Giris_Doom_Oynayan_Ajan_Gelistirme/.gitignore:
--------------------------------------------------------------------------------
1 | # Byte-compiled / optimized / DLL files
2 | __pycache__/
3 | *.py[cod]
4 | *$py.class
5 |
6 | # C extensions
7 | *.so
8 |
9 | # Distribution / packaging
10 | .Python
11 | build/
12 | develop-eggs/
13 | dist/
14 | downloads/
15 | eggs/
16 | .eggs/
17 | lib/
18 | lib64/
19 | parts/
20 | sdist/
21 | var/
22 | wheels/
23 | *.egg-info/
24 | .installed.cfg
25 | *.egg
26 | MANIFEST
27 |
28 | # PyInstaller
29 | # Usually these files are written by a python script from a template
30 | # before PyInstaller builds the exe, so as to inject date/other infos into it.
31 | *.manifest
32 | *.spec
33 |
34 | # Installer logs
35 | pip-log.txt
36 | pip-delete-this-directory.txt
37 |
38 | # Unit test / coverage reports
39 | htmlcov/
40 | .tox/
41 | .coverage
42 | .coverage.*
43 | .cache
44 | nosetests.xml
45 | coverage.xml
46 | *.cover
47 | .hypothesis/
48 | .pytest_cache/
49 |
50 | # Translations
51 | *.mo
52 | *.pot
53 |
54 | # Django stuff:
55 | *.log
56 | local_settings.py
57 | db.sqlite3
58 |
59 | # Flask stuff:
60 | instance/
61 | .webassets-cache
62 |
63 | # Scrapy stuff:
64 | .scrapy
65 |
66 | # Sphinx documentation
67 | docs/_build/
68 |
69 | # PyBuilder
70 | target/
71 |
72 | # Jupyter Notebook
73 | .ipynb_checkpoints
74 |
75 | # pyenv
76 | .python-version
77 |
78 | # celery beat schedule file
79 | celerybeat-schedule
80 |
81 | # SageMath parsed files
82 | *.sage.py
83 |
84 | # Environments
85 | .env
86 | .venv
87 | env/
88 | venv/
89 | ENV/
90 | env.bak/
91 | venv.bak/
92 |
93 | # Spyder project settings
94 | .spyderproject
95 | .spyproject
96 |
97 | # Rope project settings
98 | .ropeproject
99 |
100 | # mkdocs documentation
101 | /site
102 |
103 | # mypy
104 | .mypy_cache/
105 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Derin_Pekistirmeli_Ogrenmeye_Giris_Doom_Oynayan_Ajan_Gelistirme/README.md:
--------------------------------------------------------------------------------
1 | # DeepCon18 Deep RL Atolyesi
2 | 05-06 Ekim tarihlerinde gerçekleşen DeepCon konferansındaki Derin Pekiştirmeli Öğrenme Atölyesi
3 | Kaynak: https://github.com/FurkanArslan/deepcon-deep-rl-workshop
4 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Derin_Pekistirmeli_Ogrenmeye_Giris_Doom_Oynayan_Ajan_Gelistirme/_config.yml:
--------------------------------------------------------------------------------
1 | theme: jekyll-theme-cayman
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Derin_Pekistirmeli_Ogrenmeye_Giris_Doom_Oynayan_Ajan_Gelistirme/ayarlar/basic.cfg:
--------------------------------------------------------------------------------
1 | # Lines starting with # are treated as comments (or with whitespaces+#).
2 | # It doesn't matter if you use capital letters or not.
3 | # It doesn't matter if you use underscore or camel notation for keys, e.g. episode_timeout is the same as episodeTimeout.
4 |
5 | doom_scenario_path = basic.wad
6 | doom_map = map01
7 |
8 | # Rewards
9 | living_reward = -1
10 |
11 | # Rendering options
12 | screen_resolution = RES_320X240
13 | screen_format = GRAY8
14 | render_hud = True
15 | render_crosshair = false
16 | render_weapon = true
17 | render_decals = false
18 | render_particles = false
19 | window_visible = false
20 |
21 | # make episodes start after 20 tics (after unholstering the gun)
22 | episode_start_time = 14
23 |
24 | # make episodes finish after 300 actions (tics)
25 | episode_timeout = 300
26 |
27 | # Available buttons
28 | available_buttons =
29 | {
30 | MOVE_LEFT
31 | MOVE_RIGHT
32 | ATTACK
33 | }
34 |
35 | # Game variables that will be in the state
36 | available_game_variables = { AMMO2}
37 |
38 | mode = PLAYER
39 | doom_skill = 5
40 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Derin_Pekistirmeli_Ogrenmeye_Giris_Doom_Oynayan_Ajan_Gelistirme/models/checkpoint:
--------------------------------------------------------------------------------
1 | model_checkpoint_path: "model.ckpt"
2 | all_model_checkpoint_paths: "model.ckpt"
3 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Derin_Pekistirmeli_Ogrenmeye_Giris_Doom_Oynayan_Ajan_Gelistirme/models/model.ckpt.data-00000-of-00001:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Derin_Pekistirmeli_Ogrenmeye_Giris_Doom_Oynayan_Ajan_Gelistirme/models/model.ckpt.data-00000-of-00001
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Derin_Pekistirmeli_Ogrenmeye_Giris_Doom_Oynayan_Ajan_Gelistirme/models/model.ckpt.index:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Derin_Pekistirmeli_Ogrenmeye_Giris_Doom_Oynayan_Ajan_Gelistirme/models/model.ckpt.index
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Derin_Pekistirmeli_Ogrenmeye_Giris_Doom_Oynayan_Ajan_Gelistirme/models/model.ckpt.meta:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Derin_Pekistirmeli_Ogrenmeye_Giris_Doom_Oynayan_Ajan_Gelistirme/models/model.ckpt.meta
--------------------------------------------------------------------------------
/Atolye_Dosyalari/E-Ticaret_ve_Yapay_Zeka_Uygulamalari/E-Ticaret ve Yapay Zeka Uygulamaları_CüneytAksakallı.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/E-Ticaret_ve_Yapay_Zeka_Uygulamalari/E-Ticaret ve Yapay Zeka Uygulamaları_CüneytAksakallı.pdf
--------------------------------------------------------------------------------
/Atolye_Dosyalari/E-Ticaret_ve_Yapay_Zeka_Uygulamalari/README.md:
--------------------------------------------------------------------------------
1 | # DeepCon18 / E-Ticaret ve Yapay Zeka Uygulamalari
2 | http://derindelimavi.blogspot.com/2018/10/deepcon18-payment-anomaly-workshop.html
3 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/E-Ticaret_ve_Yapay_Zeka_Uygulamalari/best_comp.h5:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/E-Ticaret_ve_Yapay_Zeka_Uygulamalari/best_comp.h5
--------------------------------------------------------------------------------
/Atolye_Dosyalari/E-Ticaret_ve_Yapay_Zeka_Uygulamalari/circular_normalization.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/E-Ticaret_ve_Yapay_Zeka_Uygulamalari/circular_normalization.png
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Göktuğ İslamoğlu DeepCon 18.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Göktuğ İslamoğlu DeepCon 18.pdf
--------------------------------------------------------------------------------
/Atolye_Dosyalari/HalilİbrahimÇelenli_DeepCon.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/HalilİbrahimÇelenli_DeepCon.pdf
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Mamografi_Goruntuleri_Kullanarak_Meme_Kanseri_Teshisinin_Performans_Degerlendirmesi/ReadMe.md:
--------------------------------------------------------------------------------
1 |
2 | ## Mamografi Görüntüleri Kullanarak Meme Kanseri Teşhisinin Performans Değerlendirmesi
3 | ### Burcu Bektaş
4 | DeepCon'18 Atölye Çalışması
5 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Mamografi_Goruntuleri_Kullanarak_Meme_Kanseri_Teshisinin_Performans_Degerlendirmesi/lablenumeric.xlsx:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Mamografi_Goruntuleri_Kullanarak_Meme_Kanseri_Teshisinin_Performans_Degerlendirmesi/lablenumeric.xlsx
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Mamografi_Goruntuleri_Kullanarak_Meme_Kanseri_Teshisinin_Performans_Degerlendirmesi/mat2arff.m:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Mamografi_Goruntuleri_Kullanarak_Meme_Kanseri_Teshisinin_Performans_Degerlendirmesi/mat2arff.m
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Mamografi_Goruntuleri_Kullanarak_Meme_Kanseri_Teshisinin_Performans_Degerlendirmesi/onisleme.m:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Mamografi_Goruntuleri_Kullanarak_Meme_Kanseri_Teshisinin_Performans_Degerlendirmesi/onisleme.m
--------------------------------------------------------------------------------
/Atolye_Dosyalari/OpenAI_Gym_ile_Pekistirmeli_Ogrenme/README.md:
--------------------------------------------------------------------------------
1 | # DeepCon'18 Deep Reinforcement Learning Workshop
2 |
3 |
4 | ## Katılım için gereksinimler
5 |
6 | Workshop katılımı için temel Python bilgisi gerekmektedir. İnteraktif bir çalışma olacağı için laptopunuzla beraber gelmelisiniz, en geç 30 eylül tarihine kadar hazır olacak bu repoda DQN ajanının kullanım için hazır bir sınıfını bulacaksınız. Bu sınıfı openai gym cartpole ortamında nasıl kullanacağımızı ve bu sınıfın detaylarını birlikte inceleyeceğiz . [Python](https://www.python.org/downloads/) adresinden indirebileceğiniz Python'un son versiyonu bilgisayarınızda kurulu olmalıdır. Editör olarak [Pycharm](https://www.jetbrains.com/pycharm/) kullanacağım. Bu aşamalar tamamlandıktan sonra aşağıdaki kodu bilgisayarınızda çalıştırmanız zaman kazanmamız açısından önemlidir.
7 | ```
8 | git clone https://github.com/umutcanaltin/deepcon-drl-workshop.git
9 | pip install numpy matplotlib gym keras
10 | ```
11 |
12 | Görüşmek üzere!
13 | Umut Can ALTIN
14 | https://umutcanaltin.github.io/
15 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/OpenAI_Gym_ile_Pekistirmeli_Ogrenme/cartpole_örnek_hareketler.py:
--------------------------------------------------------------------------------
1 | import gym
2 | env = gym.make('CartPole-v0')
3 |
4 | env.reset()
5 | rewards = []
6 |
7 | episodes=10
8 | max_time=100
9 | for e in range(episodes):
10 | env.render()
11 | for t in range(max_time) :
12 | state, reward, done, info = env.step(env.action_space.sample()) # rastgele hareket et
13 |
14 | if done:
15 | rewards.append(t)
16 | env.reset()
17 |
18 |
19 |
20 | print(rewards)
21 | env.close()
--------------------------------------------------------------------------------
/Atolye_Dosyalari/OpenAI_Gym_ile_Pekistirmeli_Ogrenme/model.h5:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/OpenAI_Gym_ile_Pekistirmeli_Ogrenme/model.h5
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Pekistirmeli_Ogrenmede_Cok_Kollu_Haydutlar_Problemi/.gitignore:
--------------------------------------------------------------------------------
1 | # Byte-compiled / optimized / DLL files
2 | __pycache__/
3 | *.py[cod]
4 | *$py.class
5 |
6 | # C extensions
7 | *.so
8 |
9 | # Distribution / packaging
10 | .Python
11 | build/
12 | develop-eggs/
13 | dist/
14 | downloads/
15 | eggs/
16 | .eggs/
17 | lib/
18 | lib64/
19 | parts/
20 | sdist/
21 | var/
22 | wheels/
23 | *.egg-info/
24 | .installed.cfg
25 | *.egg
26 | MANIFEST
27 |
28 | # PyInstaller
29 | # Usually these files are written by a python script from a template
30 | # before PyInstaller builds the exe, so as to inject date/other infos into it.
31 | *.manifest
32 | *.spec
33 |
34 | # Installer logs
35 | pip-log.txt
36 | pip-delete-this-directory.txt
37 |
38 | # Unit test / coverage reports
39 | htmlcov/
40 | .tox/
41 | .nox/
42 | .coverage
43 | .coverage.*
44 | .cache
45 | nosetests.xml
46 | coverage.xml
47 | *.cover
48 | .hypothesis/
49 | .pytest_cache/
50 |
51 | # Translations
52 | *.mo
53 | *.pot
54 |
55 | # Django stuff:
56 | *.log
57 | local_settings.py
58 | db.sqlite3
59 |
60 | # Flask stuff:
61 | instance/
62 | .webassets-cache
63 |
64 | # Scrapy stuff:
65 | .scrapy
66 |
67 | # Sphinx documentation
68 | docs/_build/
69 |
70 | # PyBuilder
71 | target/
72 |
73 | # Jupyter Notebook
74 | .ipynb_checkpoints
75 |
76 | # IPython
77 | profile_default/
78 | ipython_config.py
79 |
80 | # pyenv
81 | .python-version
82 |
83 | # celery beat schedule file
84 | celerybeat-schedule
85 |
86 | # SageMath parsed files
87 | *.sage.py
88 |
89 | # Environments
90 | .env
91 | .venv
92 | env/
93 | venv/
94 | ENV/
95 | env.bak/
96 | venv.bak/
97 |
98 | # Spyder project settings
99 | .spyderproject
100 | .spyproject
101 |
102 | # Rope project settings
103 | .ropeproject
104 |
105 | # mkdocs documentation
106 | /site
107 |
108 | # mypy
109 | .mypy_cache/
110 | .dmypy.json
111 | dmypy.json
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Pekistirmeli_Ogrenmede_Cok_Kollu_Haydutlar_Problemi/Pekiştirmeli Öğrenmede Çok Kollu Haydutlar Problemi.pptx:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/Pekistirmeli_Ogrenmede_Cok_Kollu_Haydutlar_Problemi/Pekiştirmeli Öğrenmede Çok Kollu Haydutlar Problemi.pptx
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Pekistirmeli_Ogrenmede_Cok_Kollu_Haydutlar_Problemi/README.md:
--------------------------------------------------------------------------------
1 | # DeepCon - Pekiştirmeli Öğrenmede Çok Kollu Haydutlar Problemi
2 |
3 | ## Katılım için gereksinimler
4 |
5 | Workshop katılımı için temel Python bilgisi gerekmektedir. İnteraktif bir çalışma olacağı için laptopunuzla beraber gelmelisiniz. [Python](https://www.python.org/downloads/) adresinden indirebileceğiniz Python'un son versiyonu bilgisayarınızda kurulu olmalıdır. Bu aşamalar tamamlandıktan sonra aşağıdaki kodu bilgisayarınızda çalıştırmanız zaman kazanmamız açısından önemlidir. Kod yazımı kolaylığı için [Visual Studio Code](https://code.visualstudio.com/download) kullanmanızı öneriyoruz. Bu repository'yi alabilmek için [Git](https://git-scm.com/downloads) indirmeniz gerekmektedir.
6 |
7 | ```
8 | git clone https://github.com/kivancguckiran/deepcon-rl-workshop
9 | pip install numpy matplotlib
10 | ```
11 |
12 | Görüşmek üzere!
13 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Pekistirmeli_Ogrenmede_Cok_Kollu_Haydutlar_Problemi/min_util.py:
--------------------------------------------------------------------------------
1 | import numpy as np
2 | import matplotlib.pyplot as plt
3 |
4 |
5 | class bandit:
6 | def __init__(self, eps=0, step_size=0., initial=0, variance=1):
7 | self.epsilon = eps
8 | self.step_size = step_size
9 | self.k_arm = 5
10 | self.variance = variance
11 |
12 | self.q_true = np.asarray([-0.25, 2, 1, -1.5, 3])
13 | self.q_estimates = np.zeros(self.k_arm) + initial
14 |
15 | self.actions_taken = np.zeros(self.k_arm)
16 |
17 | def action(self):
18 | if np.random.rand() < self.epsilon:
19 | return np.random.randint(self.k_arm)
20 | else:
21 | return np.argmax(self.q_estimates)
22 |
23 | def step(self):
24 | idx = self.action()
25 |
26 | self.actions_taken[idx] += 1
27 |
28 | reward = self.q_true[idx] + np.random.randn() * self.variance
29 |
30 | if self.step_size == 0:
31 | self.q_estimates[idx] = self.q_estimates[idx] + (1.0 / self.actions_taken[idx]) * (reward - self.q_estimates[idx])
32 | else:
33 | self.q_estimates[idx] = self.q_estimates[idx] + self.step_size * (reward - self.q_estimates[idx])
34 |
35 | return reward
36 |
37 | def change_q_true(self):
38 | self.q_true = np.asarray([1, -0.5, -2, 2, 0.25])
39 |
40 | def take_steps(self, count):
41 | for _ in np.arange(count):
42 | self.step()
43 |
44 | self.showdown()
45 |
46 | def plot(self):
47 | plt.violinplot(positions=np.arange(self.k_arm), dataset=self.q_true + np.random.randn(100,self.k_arm) * self.variance)
48 | plt.plot(self.q_estimates, color='red', marker='o', linestyle='', markersize=5)
49 | plt.show()
50 |
51 | def dump(self):
52 | for i in np.arange(self.k_arm):
53 | print('Arm #%u: %u times.' % (i, self.actions_taken[i]))
54 |
55 | print('Best Arm: #%u.' % (np.argmax(self.q_true)))
56 |
57 | def showdown(self):
58 | self.dump()
59 | self.plot()
60 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Pekistirmeli_Ogrenmede_Cok_Kollu_Haydutlar_Problemi/r.md:
--------------------------------------------------------------------------------
1 |
2 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/Pekistirmeli_Ogrenmede_Cok_Kollu_Haydutlar_Problemi/util.py:
--------------------------------------------------------------------------------
1 | import numpy as np
2 | import matplotlib.pyplot as plt
3 |
4 |
5 | class bandit:
6 | def __init__(self, k_arm=10, eps=0, step_size=0.1):
7 | self.k_arm = k_arm
8 | self.epsilon = eps
9 | self.step_size = step_size
10 |
11 | self.q_true = np.random.randn(self.k_arm)
12 | self.q_estimates = np.zeros(self.k_arm)
13 |
14 | self.best_arm = np.argmax(self.q_true)
15 |
16 | def action(self):
17 | if np.random.rand() < self.epsilon:
18 | return np.random.randint(self.k_arm)
19 |
20 | return np.argmax(self.q_estimates)
21 |
22 | def step(self):
23 | idx = self.action()
24 |
25 | reward = self.q_true[idx] + np.random.randn()
26 |
27 | self.q_estimates[idx] += self.step_size * (reward - self.q_estimates[idx])
28 |
29 | return reward
30 |
31 | def take_steps(self, count):
32 | for _ in np.arange(count):
33 | self.step()
34 |
35 | self.plot_current_estimates()
36 |
37 | def plot_current_estimates(self):
38 | plt.violinplot(positions=np.arange(self.k_arm), dataset=self.q_true + np.random.randn(100,10))
39 | plt.plot(self.q_estimates, color='red', marker='o', linestyle='', markersize=5)
40 | plt.show()
41 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/ReadMe.md:
--------------------------------------------------------------------------------
1 | ### DeepCon'18 Yapay Zeka Konferansı
2 | 📚💻Atölye çalışma dosyalarına [buradan](https://github.com/deeplearningturkiye/DeepCon18/blob/master/README.md) ulaşabilirsiniz. 🎯
3 |
4 |
5 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/README.md:
--------------------------------------------------------------------------------
1 | # DeepCon'18 - Atölye - TensorFlow.js
2 | [Yapay Zeka Konferansı (DeepCon'18)](http://deepcon.deeplearningturkiye.com/) **"TensorFlow.js ile Derin Öğrenme Web Uygulaması Geliştirme"** Atölye Sunum ve Demo Kaynak Kodları
3 |
4 | [](http://deepcon.deeplearningturkiye.com/)
5 |
6 | ## Online Demo
7 | Atölye uygulamasını online olarak test etmek için [tıklayınız](https://yavuzkomecoglu.github.io/deepcon_demo/).
8 |
9 | ## Sunum
10 | - PowerPoint versiyonu için [tıklayınız.](https://github.com/yavuzKomecoglu/deepcon18_atolye_tensorflowJS/blob/master/DeepCon18_sunum.ppt)
11 | - PDF versiyonu için [tıklayınız.](https://github.com/yavuzKomecoglu/deepcon18_atolye_tensorflowJS/blob/master/DeepCon18_sunum.pdf)
12 |
13 | ## Faydalı Kaynaklar
14 | ##### Resmi Kaynaklar
15 |
16 | - [Resmi TensorFlow.js Sitesi](https://js.tensorflow.org/)
17 | - [TensorFlow.js Örnekleri](https://github.com/tensorflow/tfjs-examples)
18 | - [TensorFlow.js Giriş](https://medium.com/tensorflow/introducing-tensorflow-js-machine-learning-in-javascript-bf3eab376db)
19 | - [Yapay Sinir Ağları Görselleştirme Demosu](https://playground.tensorflow.org)
20 |
21 | ##### Diğer Kaynaklar
22 | - [Cheetsheet](https://towardsdatascience.com/50-tensorflow-js-api-explained-in-5-minutes-tensorflow-js-cheetsheet-4f8c7f9cc8b2)
23 | - [Deep Learning Türkiye Blog - TensorFlow.js ile Postür Tahmini için Derin Öğrenme – Sıddık Açıl](https://medium.com/deep-learning-turkiye/tensorflow-js-ile-post%C3%BCr-tahmini-i%C3%A7in-derin-%C3%B6%C4%9Frenme-l-f040de7355d8)
24 |
25 | ### İletişim
26 | Sosyal medya hesaplarım ve bloguma [yavuzkomecoglu.com](http://yavuzkomecoglu.com/) üzerinden ulaşabilirsiniz.
27 | Her türlü soru ve iletişim için: [komecoglu.yavuz@gmail.com](mailto:komecoglu.yavuz@gmail.com)
28 |
29 |
30 |
31 |
--------------------------------------------------------------------------------
/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/YavuzKomecoglu_DeepCon18_Sunum.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/YavuzKomecoglu_DeepCon18_Sunum.pdf
--------------------------------------------------------------------------------
/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/YavuzKomecoglu_DeepCon18_Sunum.ppt:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/YavuzKomecoglu_DeepCon18_Sunum.ppt
--------------------------------------------------------------------------------
/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/images/angry.gif:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/images/angry.gif
--------------------------------------------------------------------------------
/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/images/excited.gif:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/images/excited.gif
--------------------------------------------------------------------------------
/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/images/laugh.gif:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/images/laugh.gif
--------------------------------------------------------------------------------
/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/images/out.gif:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/images/out.gif
--------------------------------------------------------------------------------
/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/images/sad.gif:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/images/sad.gif
--------------------------------------------------------------------------------
/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/images/sleep.gif:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/images/sleep.gif
--------------------------------------------------------------------------------
/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/sounds/shutter.mp3:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/sounds/shutter.mp3
--------------------------------------------------------------------------------
/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/sounds/shutter.ogg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Atolye_Dosyalari/TensorFlowJS_Derin_Ogrenme_Web Uygulamasi_Gelistirme/demo/assets/sounds/shutter.ogg
--------------------------------------------------------------------------------
/DEEPCON18.jpeg:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/DEEPCON18.jpeg
--------------------------------------------------------------------------------
/Konusmaci_Sunumlari/1-DeepCon_AykutErdem.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Konusmaci_Sunumlari/1-DeepCon_AykutErdem.pdf
--------------------------------------------------------------------------------
/Konusmaci_Sunumlari/2-DeepCon_Onur Koc.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Konusmaci_Sunumlari/2-DeepCon_Onur Koc.pdf
--------------------------------------------------------------------------------
/Konusmaci_Sunumlari/3-DeepCon_Av.SertalSiraci.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Konusmaci_Sunumlari/3-DeepCon_Av.SertalSiraci.pdf
--------------------------------------------------------------------------------
/Konusmaci_Sunumlari/4-DeepCon_AliOsmanOrs.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Konusmaci_Sunumlari/4-DeepCon_AliOsmanOrs.pdf
--------------------------------------------------------------------------------
/Konusmaci_Sunumlari/5-DeepCon_ESerdarGokpinar.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Konusmaci_Sunumlari/5-DeepCon_ESerdarGokpinar.pdf
--------------------------------------------------------------------------------
/Konusmaci_Sunumlari/6-DeepCon_BerkSunar.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Konusmaci_Sunumlari/6-DeepCon_BerkSunar.pdf
--------------------------------------------------------------------------------
/Konusmaci_Sunumlari/7-DeepCon_AlperGerçek.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Konusmaci_Sunumlari/7-DeepCon_AlperGerçek.pdf
--------------------------------------------------------------------------------
/Konusmaci_Sunumlari/9-DeepCon_ErdemYoruk.pdf:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/Konusmaci_Sunumlari/9-DeepCon_ErdemYoruk.pdf
--------------------------------------------------------------------------------
/Konusmaci_Sunumlari/ReadMe.md:
--------------------------------------------------------------------------------
1 |
2 | ## DeepCon'18 Yapay Zeka Konferansı
3 | ### 5 Ekim 2018 tarihinde yapılan konuşmaların sunum dosyalarına bu dosyadan ulaşabilirsiniz.
4 |
--------------------------------------------------------------------------------
/atolye_egitmenleri1.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/atolye_egitmenleri1.png
--------------------------------------------------------------------------------
/atolye_egitmenleri2.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/atolye_egitmenleri2.png
--------------------------------------------------------------------------------
/atolye_egitmenleri3.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/atolye_egitmenleri3.png
--------------------------------------------------------------------------------
/deepcon_logo.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/deeplearningturkiye/DeepCon18/61ecdd8c19dd054a6c5b83b77b6ae48d3790cf8e/deepcon_logo.png
--------------------------------------------------------------------------------