#N canvas 871 74 498 738 10;
#X obj 28 237 gemwin;
#X msg 28 211 create \, 1;
#N canvas 463 0 765 790 pix2sig_stuff~ 0;
#X obj 120 35 gemhead;
#X obj 120 132 pix_texture;
#X obj 119 274 outlet~;
#X obj 139 185 square 4;
#X obj 139 163 separator;
#X obj 61 165 separator;
#X obj 120 101 pix_video;
#X msg 186 64 dimen 640 480;
#X obj 26 36 block~ 2048;
#X msg 186 38 dimen 320 240;
#X msg 76 535 getprecision;
#X msg 93 696 getlearnrate;
#X msg 65 671 learnrate 0.2;
#X msg 424 459 getneurons;
#X msg 404 206 train;
#X obj 31 227 inlet~;
#X msg 65 647 learnrate 0.05;
#X text 296 49 <- input dimension;
#X msg 76 498 precision \$1;
#X floatatom 76 481 5 0 0 0 - - -;
#X text 42 335 precision:;
#X text 53 358 1: means every pixel is used in calculation;
#X text 53 372 2: only every second pixel;
#X text 53 386 ...;
#X obj 62 411 loadbang;
#X msg 407 401 neurons 2048;
#X msg 407 422 neurons 64;
#X text 403 336 neurons:;
#X text 416 357 nr. of neurons used in the calculation;
#X text 415 370 (_MUST_ be the same as the buffersize !!!);
#X text 43 615 learnrate:;
#X msg 62 456 precision 1;
#X msg 62 436 precision 4;
#X text 397 126 train:;
#X text 417 152 trains the neural net;
#X text 418 166 (the current video frame to;
#X text 425 178 the current audio block);
#X obj 61 252 pix_linNN;
#X text 346 592 save/load;
#X text 359 614 saves/load the actual trained net to/from a file;
#X msg 440 684 load net.dat;
#X msg 440 664 save net.dat;
#X obj 78 226 r \$0-linNN;
#X obj 404 233 s \$0-linNN;
#X obj 62 564 s \$0-linNN;
#X obj 407 492 s \$0-linNN;
#X obj 65 725 s \$0-linNN;
#X obj 440 723 s \$0-linNN;
#X connect 0 0 6 0;
#X connect 1 0 4 0;
#X connect 1 0 5 0;
#X connect 4 0 3 0;
#X connect 5 0 37 0;
#X connect 6 0 1 0;
#X connect 7 0 6 0;
#X connect 9 0 6 0;
#X connect 10 0 44 0;
#X connect 11 0 46 0;
#X connect 12 0 46 0;
#X connect 13 0 45 0;
#X connect 14 0 43 0;
#X connect 15 0 37 0;
#X connect 16 0 46 0;
#X connect 18 0 44 0;
#X connect 19 0 18 0;
#X connect 24 0 32 0;
#X connect 25 0 45 0;
#X connect 26 0 45 0;
#X connect 31 0 44 0;
#X connect 32 0 44 0;
#X connect 37 1 2 0;
#X connect 40 0 47 0;
#X connect 41 0 47 0;
#X connect 42 0 37 0;
#X restore 87 492 pd pix2sig_stuff~;
#X msg 102 212 0 \, destroy;
#X obj 114 537 unsig~;
#X obj 204 382 osc~ 440;
#X obj 203 406 *~;
#X obj 235 406 tgl 15 0 empty empty empty 0 -6 0 8 -262144 -1 -1 0
1;
#X obj 205 446 sig~ 0;
#X floatatom 115 558 8 0 0 0 - - -;
#X text 199 230 <- create gemwin;
#X obj 39 392 readsf~;
#X obj 39 351 openpanel;
#X msg 39 371 open \$1;
#X obj 39 330 bng 15 250 50 0 empty empty empty 0 -6 0 8 -262144 -1
-1;
#X text 65 329 <- load sample for training;
#X obj 120 367 tgl 25 0 empty empty empty 0 -6 0 8 -195568 -1 -1 0
1;
#X floatatom 204 364 5 0 0 0 - - -;
#X text 270 381 <- simple osc for training;
#X text 260 447 <- to train silence;
#X obj 83 413 bng 15 250 50 0 empty empty empty 0 -6 0 8 -262144 -1
-1;
#X text 214 491 <- audio/video work;
#X obj 88 634 dac~;
#X obj 88 609 *~;
#X obj 116 609 dbtorms;
#X floatatom 116 591 5 0 0 0 - - -;
#X text 166 588 <- outvol in dB;
#X text 110 703 Georg Holzmann <grh@mur.at> \, 2004;
#X text 24 23 pix_linNN:;
#X text 22 58 (see also pix_recNN !!!);
#X text 24 90 pix_linNN~ calculates an audio signal out of a video
frame with a linear neural network \, which can be trained.;
#X text 24 124 The network has one neuron per audio sample: this neuron
has three inputs (a RGB-signal) \, a weight vector for each of the
inputs \, a bias value and a linear output function.;
#X connect 1 0 0 0;
#X connect 2 0 4 0;
#X connect 2 0 23 0;
#X connect 3 0 0 0;
#X connect 4 0 9 0;
#X connect 5 0 6 0;
#X connect 6 0 2 0;
#X connect 7 0 6 1;
#X connect 8 0 2 0;
#X connect 11 0 2 0;
#X connect 11 1 20 0;
#X connect 12 0 13 0;
#X connect 13 0 11 0;
#X connect 14 0 12 0;
#X connect 16 0 11 0;
#X connect 17 0 5 0;
#X connect 23 0 22 0;
#X connect 23 0 22 1;
#X connect 24 0 23 1;
#X connect 25 0 24 0;