As explained in 00-irisxl Deep Learning has 2 pilars : Neural Network and Convolution -This appli still demonstates a Neural Network. Method is exactly similar to the one used in 00-irisxl. -But now we deal with an Input of 784 pos (=28x28) to represent one hand-written digit - Incidentally this example is directly derived from the first application which launched the Deep Learning In addition this application shows how to use powerful GRID and PIXEL functions. Much grain to grind !!!data:image/png;base64,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As explained in 00-irisxl Deep Learning has 2 pilars : Neural Network and Convolution -This appli still demonstates a Neural Network. Method is exactly similar to the one used in 00-irisxl. -But now we deal with an Input of 784 pos (=28x28) to represent one hand-written digit - Incidentally this example is directly derived from the first application which launched the Deep Learning In addition this application shows how to use powerful GRID and PIXEL functions. Much grain to grind !!!

3

Update of a matrix by a "modification" vector defined by bwd 2 for w2 and bwd 3 for w3

Add numbers to the right side of matrix. Use RDLIS to create a random vector + filter ADR

234

234

reports a list of all pixels in the given costume. Each pixel is represented by a 4-item sublist of RGBA values

pixelspixels width heightheightcostume var i, pixels = [], src = costume.contents.getContext('2d').getImageData( 0, 0, costume.width(), costume.height() ); for (i = 0; i < src.data.length; i += 4) { pixels.push(new List([ src.data[i], src.data[i + 1], src.data[i + 2], src.data[i + 3] ])); } return new List(pixels);overwrite the pixels in the given costume with another list of pixels. Caution! This operation directly changes the costume. Make sure to use a copy in case you would like to revert changes later.

reports the sprite's actual current costume, which may or may not be part of its wardrobe e.g. if it is a copy

de:für _ = _ bis _ _ 110i

newProvides LOOP as a function of one input that runs the body of the LET with A set to the function input, so the body can run itself recursively. See COPY block in Variables for an example of use.

new valuecopy VALUE N times reports a list containing N (identical) copies of VALUE

ypncnl

xy

w-2h-200

hsclpa

display the given bitmap in a copy of the current costume (so you can switch back to the original costume again)

pt:mostra a imagem _ takes a snapshot with the webcam and reports it as a new costume, or zero if the user cancels

pt:$camera a imagem actual do vídeo

data:image/png;base64,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,0,0,1,0,0,0,0,0,024.79193355453274330.34119510000.15,0.17,0.64,0.04,0.05,0.16,0.04,0.04,0.05,0.121,0,10,1,00,0,101rgb(0,0,0)0.01,0.06,0.073,0.052,0.08,0.018,0.072,0.098,0.038,0.026,0.055,0.042,0.02,0.079,0.026,0.05,0.056,0.096,0.096,0.049,0.094,0.063,0.089,0.014,0.088,0.094,0.064,0.087,0.097,0.041,0.06,0.073,0.059,0.084,0.07,0.083,0.08,0.089,0.06,0.0380.098,0.052,0.078,0.063,0.021,0.065,0.085,0.098,0.084,0.019,0.017,0.091,0.044,0.051,0.06,0.07,0.045,0.08,0.065,0.086,0.093,0.058,0.068,0.04,0.073,0.018,0.025,0.092,0.032,0.031,0.081,0.093,0.029,0.095,0.013,0.043,0.082,0.087,0.019,0.0550.075,0.047,0.046,0.019,0.088,0.079,0.056,0.086,0.02,0.092,0.044,0.061,0.079,0.028,0.046,0.015,0.067,0.018,0.016,0.044,0.012,0.067,0.094,0.067,0.055,0.024,0.064,0.089,0.066,0.087,0.089,0.06,0.063,0.077,0.029,0.081,0.019,0.057,0.024,0.0210.096,0.08,0.064,0.069,0.048,0.022,0.08,0.026,0.086,0.084,0.098,0.033,0.075,0.086,0.086,0.034,0.087,0.086,0.082,0.075,0.054,0.046,0.03,0.071,0.025,0.038,0.082,0.067,0.035,0.045,0.03,0.03,0.026,0.047,0.033,0.05,0.09,0.012,0.019,0.0270.014,0.091,0.034,0.046,0.054,0.065,0.037,0.056,0.06,0.079,0.059,0.026,0.08,0.013,0.079,0.097,0.029,0.024,0.044,0.058,0.072,0.077,0.011,0.031,0.091,0.073,0.077,0.053,0.033,0.025,0.095,0.049,0.013,0.028,0.088,0.072,0.01,0.061,0.068,0.0790.067,0.098,0.019,0.028,0.097,0.096,0.049,0.046,0.038,0.086,0.075,0.096,0.01,0.086,0.082,0.048,0.099,0.09,0.078,0.03,0.012,0.061,0.091,0.078,0.05,0.021,0.019,0.02,0.046,0.081,0.048,0.016,0.078,0.082,0.029,0.025,0.099,0.041,0.091,0.0680.065,0.018,0.074,0.081,0.028,0.014,0.073,0.031,0.034,0.05,0.063,0.052,0.071,0.088,0.033,0.04,0.034,0.033,0.022,0.074,0.082,0.016,0.031,0.075,0.029,0.064,0.065,0.064,0.064,0.033,0.087,0.063,0.068,0.062,0.096,0.044,0.065,0.051,0.033,0.0880.044,0.067,0.044,0.095,0.047,0.045,0.08,0.084,0.04,0.079,0.091,0.058,0.059,0.01,0.082,0.082,0.03,0.093,0.074,0.093,0.087,0.098,0.029,0.066,0.011,0.07,0.019,0.02,0.02,0.093,0.089,0.048,0.02,0.017,0.09,0.042,0.084,0.042,0.029,0.0190.018,0.012,0.076,0.053,0.083,0.098,0.061,0.063,0.013,0.068,0.097,0.035,0.022,0.036,0.089,0.047,0.029,0.046,0.098,0.09,0.041,0.077,0.089,0.036,0.051,0.054,0.076,0.014,0.051,0.075,0.079,0.089,0.086,0.062,0.087,0.095,0.01,0.099,0.046,0.0590.022,0.085,0.042,0.091,0.046,0.077,0.049,0.099,0.057,0.043,0.064,0.054,0.03,0.075,0.088,0.028,0.042,0.087,0.084,0.093,0.047,0.015,0.022,0.073,0.087,0.012,0.065,0.097,0.067,0.087,0.071,0.011,0.023,0.045,0.085,0.045,0.023,0.09,0.043,0.02300,0,0,1,0,0,0,0,0,01,0,0,0,0,0,0,0,0,00,1,0,0,0,0,0,0,0,00,0,1,0,0,0,0,0,0,00,0,0,0,1,0,0,0,0,00,0,0,0,0,1,0,0,0,00,0,0,0,0,0,1,0,0,00,0,0,0,0,0,0,1,0,00,0,0,0,0,0,0,0,1,00,0,0,0,0,0,0,0,0,10,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,25527,27,27,255255,255,255,255250,250,250,25552,52,52,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2552,2,2,255151,151,151,255254,254,254,255254,254,254,25584,84,84,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,25576,76,76,255254,254,254,255254,254,254,255219,219,219,25530,30,30,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,25544,44,44,255239,239,239,255254,254,254,255254,254,254,25552,52,52,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,25512,12,12,255182,182,182,255254,254,254,255254,254,254,255184,184,184,25513,13,13,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,255143,143,143,255254,254,254,255254,254,254,255226,226,226,2559,9,9,2550,0,0,2550,0,0,2550,0,0,25540,40,40,255223,223,223,255230,230,230,2559,9,9,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,255135,135,135,255251,251,251,255254,254,254,255185,185,185,25514,14,14,2550,0,0,2550,0,0,2550,0,0,25542,42,42,255178,178,178,255254,254,254,255254,254,254,2559,9,9,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,255100,100,100,255251,251,251,255254,254,254,255202,202,202,2558,8,8,2550,0,0,2550,0,0,2550,0,0,25527,27,27,255185,185,185,255254,254,254,255254,254,254,255205,205,205,2555,5,5,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2558,8,8,255163,163,163,255249,249,249,255254,254,254,255254,254,254,25583,83,83,255133,133,133,25595,95,95,25595,95,95,25521,21,21,255202,202,202,255255,255,255,255254,254,254,255204,204,204,25564,64,64,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,255139,139,139,255254,254,254,255254,254,254,255254,254,254,255254,254,254,255251,251,251,255254,254,254,255254,254,254,255254,254,254,255251,251,251,255254,254,254,255254,254,254,255207,207,207,25510,10,10,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,25555,55,55,255248,248,248,255254,254,254,255254,254,254,255254,254,254,255254,254,254,255254,254,254,255254,254,254,255251,251,251,255254,254,254,255254,254,254,255254,254,254,255242,242,242,25563,63,63,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2557,7,7,255222,222,222,255254,254,254,255254,254,254,255254,254,254,255254,254,254,255221,221,221,255157,157,157,25594,94,94,25547,47,47,255254,254,254,255254,254,254,255254,254,254,255105,105,105,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,25511,11,11,255254,254,254,255254,254,254,255254,254,254,255247,247,247,255133,133,133,25518,18,18,2550,0,0,25548,48,48,255232,232,232,255254,254,254,255242,242,242,25573,73,73,2555,5,5,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2558,8,8,255159,159,159,255131,131,131,255184,184,184,25551,51,51,2550,0,0,2550,0,0,25527,27,27,255219,219,219,255254,254,254,255254,254,254,25573,73,73,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,25530,30,30,255213,213,213,255254,254,254,255242,242,242,255105,105,105,2555,5,5,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,255172,172,172,255254,254,254,255239,239,239,25563,63,63,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,255131,131,131,255253,253,253,255241,241,241,255137,137,137,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,25558,58,58,255237,237,237,255254,254,254,255124,124,124,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,255227,227,227,255254,254,254,255159,159,159,2553,3,3,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,255230,230,230,255254,254,254,25536,36,36,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,2550,0,0,255