time.1 <- Sys.time()
format(time.1, "%Y-%m-%d-%H%M%S")## [1] "2015-02-26-004309"feature.matrix is a matrix
load(file="../Wnd-Charm/plankton-train-wndchrm-features.Rdata", verbose=TRUE) Loading objects:
feature.matrixstr(feature.matrix) num [1:30336, 1:2895] 1 1 1 1 1 1 1 1 1 1 ...
- attr(*, "dimnames")=List of 2
..$ : chr [1:30336] "100224-l.sig" "100723-l.sig" "101165-l.sig" "101232-l.sig" ...
..$ : chr [1:2895] "class" "Edge Features () [0]" "Edge Features () [1]" "Edge Features () [2]" ...trainFeatures <- colnames(feature.matrix)Column 1 of the feature.matrix defines the plankton class.
counts1 <- table(feature.matrix[,1])
counts1
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
889 13 71 49 16 696 532 242 393 170 815 1934 694 77 681
16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
173 96 178 63 286 106 49 87 30 899 1189 24 201 113 42
31 32 33 34 35 36 37 38 39 40 41 42 43 44 45
53 38 55 363 394 914 519 500 36 92 80 88 385 536 96
46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
27 14 136 38 511 10 31 85 114 64 16 10 127 75 35
61 62 63 64 65 66 67 68 69 70 71 72 73 74 75
229 9 19 23 336 132 12 190 412 274 150 76 703 123 43
76 77 78 79 80 81 82 83 84 85 86 87 88 89 90
56 14 61 14 24 141 131 108 372 625 1172 113 108 13 65
91 92 93 94 95 96 97 98 99 100 101 102 103 104 105
287 158 52 49 153 174 212 135 483 179 57 247 29 30 128
106 107 108 109 110 111 112 113 114 115 116 117 118 119 120
21 24 38 708 54 1979 678 29 439 417 352 236 73 317 175
121
425 The feature.matrix row names define the image filename. For example, the original image 100224.jpg was converted to 100224.tif. The file of wndchrm properties was named 100224-l.sig, which is the rowname of feature.matrix.
skimage is a data.frame
skimage <- read.csv("../skimage/train-properties-MASTER.csv", as.is=TRUE)
str(skimage)'data.frame': 30336 obs. of 31 variables:
$ area : num 273 451 292 319 444 161 336 176 165 397 ...
$ area_convex : num 1520 4579 2022 1307 2877 ...
$ area_filled : num 273 451 292 320 446 162 337 176 166 397 ...
$ box_max_col : num 48 71 64 40 58 32 49 32 49 96 ...
$ box_max_row : num 56 80 66 64 88 40 68 71 63 79 ...
$ box_min_col : num 0 0 6 8 0 8 11 0 11 10 ...
$ box_min_row : num 8 8 10 8 11 8 11 0 2 10 ...
$ centroid_col : num 24.7 33.1 35.5 24.2 28.3 ...
$ centroid_row : num 29.8 40.6 35.1 33.7 50.6 ...
$ countCoords : num 273 451 292 319 444 161 336 176 165 397 ...
$ diameter_equivalent : num 18.6 24 19.3 20.2 23.8 ...
$ euler_number : num 0 0 0 -1 -1 -1 -1 0 -1 0 ...
$ filename : chr "100224.jpg" "100723.jpg" "101165.jpg" "101232.jpg" ...
$ inertia_tensor_eigenvalue1: num 126 331 161 199 342 ...
$ inertia_tensor_eigenvalue2: num 83.9 254.8 80.7 36.3 97.7 ...
$ label : num 1 2 1 1 1 1 1 1 2 2 ...
$ length_major_axis : num 44.9 72.8 50.7 56.4 74 ...
$ length_minor_axis : num 36.6 63.9 35.9 24.1 39.5 ...
$ moments_hu1 : num 0.768 1.298 0.827 0.737 0.99 ...
$ moments_hu2 : num 0.0236 0.0284 0.075 0.2597 0.3024 ...
$ moments_hu3 : num 0.10632 0.18037 0.00876 0.01296 0.16193 ...
$ moments_hu4 : num 0.00261 0.01568 0.00525 0.00066 0.08363 ...
$ moments_hu5 : num -8.95e-06 -3.43e-04 1.33e-05 1.81e-06 9.71e-03 ...
$ moments_hu6 : num 0.000209 0.002065 -0.000807 -0.000237 0.042954 ...
$ moments_hu7 : num -4.27e-05 -7.60e-04 3.30e-05 6.76e-07 6.33e-04 ...
$ orientation : num -0.568 0.381 0.31 1.223 1.125 ...
$ perimeter : num 175 340 217 218 315 ...
$ ratio_eccentricity : num 0.577 0.479 0.706 0.904 0.845 ...
$ ratio_extent : num 0.1185 0.0882 0.0899 0.178 0.0994 ...
$ ratio_solidity : num 0.1796 0.0985 0.1444 0.2441 0.1543 ...
$ train : int 0 0 0 0 0 0 0 0 0 0 ...head(skimage,2) area area_convex area_filled box_max_col box_max_row box_min_col
1 273 1520 273 48 56 0
2 451 4579 451 71 80 0
box_min_row centroid_col centroid_row countCoords diameter_equivalent
1 8 24.67399 29.82784 273 18.64388
2 8 33.11973 40.62084 451 23.96312
euler_number filename inertia_tensor_eigenvalue1
1 0 100224.jpg 125.8685
2 0 100723.jpg 330.8140
inertia_tensor_eigenvalue2 label length_major_axis length_minor_axis
1 83.91497 1 44.87646 36.64205
2 254.80400 2 72.75317 63.85033
moments_hu1 moments_hu2 moments_hu3 moments_hu4 moments_hu5
1 0.7684377 0.02361631 0.1063240 0.002614545 -8.948032e-06
2 1.2984877 0.02840454 0.1803685 0.015682721 -3.427971e-04
moments_hu6 moments_hu7 orientation perimeter ratio_eccentricity
1 0.0002093655 -4.266406e-05 -0.567969 175.4325 0.5773322
2 0.0020647507 -7.603912e-04 0.380896 340.2792 0.4793396
ratio_extent ratio_solidity train
1 0.11848958 0.17960526 0
2 0.08822379 0.09849312 0skimage$train defines the plankton class.
skimage$filename defines the image filename (e.g., 10024.jpg).
Filename comparisons betweem skimage and the feature.matrix must drop the extension (e.g., 10024).
The cause was not investigated, but for some reason traversing the directory tree of plankton .tif files using CentOS/IPython resulted in a slightly different directory order than traversing the corresponding directory tree of the original .jpg files.
Instead of resolving the original problem, a conversion table was created the tells the order for the two sources.
Here skimage rows are ordered to match the feature.matrix rows using filenames.
xlate <- read.csv("wndchrm-skimage-index-translate.csv", as.is=TRUE)
skiorder <- skimage$trainUse brute force here
for (i in 1:121)
{
skimage$train[skiorder == xlate$skimage[i]] <- xlate$wndchrm[i]
}wndchrm <- feature.matrix[order(feature.matrix[,1], row.names(feature.matrix) ),]
skimage <- skimage[order(skimage$train, skimage$filename), ]all((skimage$train) == feature.matrix[,1])## [1] TRUEsum((skimage$train) == feature.matrix[,1])## [1] 30336wndchrm
splits <- strsplit(row.names(feature.matrix), "-")
wndchrmFileIndex <- unlist(lapply(splits, "[", 1))
head(wndchrmFileIndex)## [1] "100224" "100723" "101165" "101232" "101260" "101358"skimage
splits <- strsplit(skimage$filename, "\\.")
skimageFileIndex <- unlist(lapply(splits, "[", 1))
all(wndchrmFileIndex == skimageFileIndex)## [1] TRUEsum(wndchrmFileIndex == skimageFileIndex)## [1] 30336Drop columns skimage\(train and skimage\)filename since they align with corresponding info in the feature.matrix.
str(skimage)'data.frame': 30336 obs. of 31 variables:
$ area : num 273 451 292 319 444 161 336 176 165 397 ...
$ area_convex : num 1520 4579 2022 1307 2877 ...
$ area_filled : num 273 451 292 320 446 162 337 176 166 397 ...
$ box_max_col : num 48 71 64 40 58 32 49 32 49 96 ...
$ box_max_row : num 56 80 66 64 88 40 68 71 63 79 ...
$ box_min_col : num 0 0 6 8 0 8 11 0 11 10 ...
$ box_min_row : num 8 8 10 8 11 8 11 0 2 10 ...
$ centroid_col : num 24.7 33.1 35.5 24.2 28.3 ...
$ centroid_row : num 29.8 40.6 35.1 33.7 50.6 ...
$ countCoords : num 273 451 292 319 444 161 336 176 165 397 ...
$ diameter_equivalent : num 18.6 24 19.3 20.2 23.8 ...
$ euler_number : num 0 0 0 -1 -1 -1 -1 0 -1 0 ...
$ filename : chr "100224.jpg" "100723.jpg" "101165.jpg" "101232.jpg" ...
$ inertia_tensor_eigenvalue1: num 126 331 161 199 342 ...
$ inertia_tensor_eigenvalue2: num 83.9 254.8 80.7 36.3 97.7 ...
$ label : num 1 2 1 1 1 1 1 1 2 2 ...
$ length_major_axis : num 44.9 72.8 50.7 56.4 74 ...
$ length_minor_axis : num 36.6 63.9 35.9 24.1 39.5 ...
$ moments_hu1 : num 0.768 1.298 0.827 0.737 0.99 ...
$ moments_hu2 : num 0.0236 0.0284 0.075 0.2597 0.3024 ...
$ moments_hu3 : num 0.10632 0.18037 0.00876 0.01296 0.16193 ...
$ moments_hu4 : num 0.00261 0.01568 0.00525 0.00066 0.08363 ...
$ moments_hu5 : num -8.95e-06 -3.43e-04 1.33e-05 1.81e-06 9.71e-03 ...
$ moments_hu6 : num 0.000209 0.002065 -0.000807 -0.000237 0.042954 ...
$ moments_hu7 : num -4.27e-05 -7.60e-04 3.30e-05 6.76e-07 6.33e-04 ...
$ orientation : num -0.568 0.381 0.31 1.223 1.125 ...
$ perimeter : num 175 340 217 218 315 ...
$ ratio_eccentricity : num 0.577 0.479 0.706 0.904 0.845 ...
$ ratio_extent : num 0.1185 0.0882 0.0899 0.178 0.0994 ...
$ ratio_solidity : num 0.1796 0.0985 0.1444 0.2441 0.1543 ...
$ train : int 1 1 1 1 1 1 1 1 1 1 ...skimage$filename <- NULL
skimage$train <- NULL
str(skimage)'data.frame': 30336 obs. of 29 variables:
$ area : num 273 451 292 319 444 161 336 176 165 397 ...
$ area_convex : num 1520 4579 2022 1307 2877 ...
$ area_filled : num 273 451 292 320 446 162 337 176 166 397 ...
$ box_max_col : num 48 71 64 40 58 32 49 32 49 96 ...
$ box_max_row : num 56 80 66 64 88 40 68 71 63 79 ...
$ box_min_col : num 0 0 6 8 0 8 11 0 11 10 ...
$ box_min_row : num 8 8 10 8 11 8 11 0 2 10 ...
$ centroid_col : num 24.7 33.1 35.5 24.2 28.3 ...
$ centroid_row : num 29.8 40.6 35.1 33.7 50.6 ...
$ countCoords : num 273 451 292 319 444 161 336 176 165 397 ...
$ diameter_equivalent : num 18.6 24 19.3 20.2 23.8 ...
$ euler_number : num 0 0 0 -1 -1 -1 -1 0 -1 0 ...
$ inertia_tensor_eigenvalue1: num 126 331 161 199 342 ...
$ inertia_tensor_eigenvalue2: num 83.9 254.8 80.7 36.3 97.7 ...
$ label : num 1 2 1 1 1 1 1 1 2 2 ...
$ length_major_axis : num 44.9 72.8 50.7 56.4 74 ...
$ length_minor_axis : num 36.6 63.9 35.9 24.1 39.5 ...
$ moments_hu1 : num 0.768 1.298 0.827 0.737 0.99 ...
$ moments_hu2 : num 0.0236 0.0284 0.075 0.2597 0.3024 ...
$ moments_hu3 : num 0.10632 0.18037 0.00876 0.01296 0.16193 ...
$ moments_hu4 : num 0.00261 0.01568 0.00525 0.00066 0.08363 ...
$ moments_hu5 : num -8.95e-06 -3.43e-04 1.33e-05 1.81e-06 9.71e-03 ...
$ moments_hu6 : num 0.000209 0.002065 -0.000807 -0.000237 0.042954 ...
$ moments_hu7 : num -4.27e-05 -7.60e-04 3.30e-05 6.76e-07 6.33e-04 ...
$ orientation : num -0.568 0.381 0.31 1.223 1.125 ...
$ perimeter : num 175 340 217 218 315 ...
$ ratio_eccentricity : num 0.577 0.479 0.706 0.904 0.845 ...
$ ratio_extent : num 0.1185 0.0882 0.0899 0.178 0.0994 ...
$ ratio_solidity : num 0.1796 0.0985 0.1444 0.2441 0.1543 ...Convert to matrix and cbind to feature.matrix
skimage <- as.matrix(skimage)
dim(feature.matrix)[1] 30336 2895dim(skimage)[1] 30336 29feature.matrix <- cbind(feature.matrix, skimage)
dim(feature.matrix)[1] 30336 2924Let’s treat the plankton class as a separate vector and remove class from the matrix
train.class <- feature.matrix[, 1]
train.features <- feature.matrix[,-1]Final look at number of images by feature
table(train.class)train.class
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
889 13 71 49 16 696 532 242 393 170 815 1934 694 77 681
16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
173 96 178 63 286 106 49 87 30 899 1189 24 201 113 42
31 32 33 34 35 36 37 38 39 40 41 42 43 44 45
53 38 55 363 394 914 519 500 36 92 80 88 385 536 96
46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
27 14 136 38 511 10 31 85 114 64 16 10 127 75 35
61 62 63 64 65 66 67 68 69 70 71 72 73 74 75
229 9 19 23 336 132 12 190 412 274 150 76 703 123 43
76 77 78 79 80 81 82 83 84 85 86 87 88 89 90
56 14 61 14 24 141 131 108 372 625 1172 113 108 13 65
91 92 93 94 95 96 97 98 99 100 101 102 103 104 105
287 158 52 49 153 174 212 135 483 179 57 247 29 30 128
106 107 108 109 110 111 112 113 114 115 116 117 118 119 120
21 24 38 708 54 1979 678 29 439 417 352 236 73 317 175
121
425 dim(train.features)[1] 30336 2923object.size(train.features)711597520 bytesobject.size(train.class)2184368 bytessave(train.features, train.class, file="plankton-train-wndchrm-skimage-features.Rdata")Save training and testing data into separate .Rdata files so testing data is only brought into memory after model has been built.
Save some memory to process testing set now
rm(train.features, train.class)load(file="../Wnd-Charm/plankton-test-wndchrm-features.Rdata", verbose=TRUE) Loading objects:
feature.matrixstr(feature.matrix) num [1:130400, 1:2898] 285 387 1999 332 967 ...
- attr(*, "dimnames")=List of 2
..$ : chr [1:130400] "1-l.sig" "10-l.sig" "100-l.sig" "1000-l.sig" ...
..$ : chr [1:2898] "Edge Features () [0]" "Edge Features () [1]" "Edge Features () [2]" "Edge Features () [3]" ...testFeatures <- colnames(feature.matrix)skimage <- read.csv("../skimage/test-properties-MASTER.csv", as.is=TRUE)
str(skimage)'data.frame': 130400 obs. of 30 variables:
$ area : num 166 216 1413 138 945 ...
$ area_convex : num 217 413 3673 241 1028 ...
$ area_filled : num 166 216 3478 138 945 ...
$ box_max_col : num 23 44 86 27 51 36 76 30 45 24 ...
$ box_max_row : num 44 32 72 35 51 35 48 32 43 39 ...
$ box_min_col : num 10 12 11 10 13 0 10 10 11 1 ...
$ box_min_row : num 12 11 10 10 11 12 8 10 11 3 ...
$ centroid_col : num 15.1 27.8 46.6 15.1 31.4 ...
$ centroid_row : num 28.3 25.1 40.6 22.6 30.9 ...
$ countCoords : num 166 216 1413 138 945 ...
$ diameter_equivalent : num 14.5 16.6 42.4 13.3 34.7 ...
$ euler_number : num 0 0 -17 0 0 -7 -8 -2 -1 0 ...
$ filename : chr "1.jpg" "10.jpg" "100.jpg" "1000.jpg" ...
$ inertia_tensor_eigenvalue1: num 73.9 52.2 445.2 34.4 122.1 ...
$ inertia_tensor_eigenvalue2: num 2.67 20.58 303.03 10.53 47.33 ...
$ label : num 3 1 1 1 1 1 1 1 2 1 ...
$ length_major_axis : num 34.4 28.9 84.4 23.5 44.2 ...
$ length_minor_axis : num 6.53 18.15 69.63 12.98 27.52 ...
$ moments_hu1 : num 0.461 0.337 0.53 0.326 0.179 ...
$ moments_hu2 : num 0.18418 0.02142 0.01012 0.02995 0.00625 ...
$ moments_hu3 : num 8.48e-04 2.84e-02 1.62e-04 1.37e-02 3.40e-06 ...
$ moments_hu4 : num 5.42e-04 1.55e-03 5.10e-05 3.58e-04 3.80e-07 ...
$ moments_hu5 : num 3.65e-07 -1.02e-05 4.50e-09 -4.20e-07 3.07e-13 ...
$ moments_hu6 : num 2.10e-04 -1.45e-04 1.71e-06 2.49e-05 1.42e-08 ...
$ moments_hu7 : num -4.92e-08 2.73e-07 1.12e-09 -6.70e-07 -3.05e-13 ...
$ orientation : num -1.365 0.327 0.397 -1.112 0.845 ...
$ perimeter : num 72.4 92.9 801.3 73.7 136.5 ...
$ ratio_eccentricity : num 0.982 0.778 0.565 0.833 0.782 ...
$ ratio_extent : num 0.399 0.321 0.304 0.325 0.622 ...
$ ratio_solidity : num 0.765 0.523 0.385 0.573 0.919 ...head(skimage,2) area area_convex area_filled box_max_col box_max_row box_min_col
1 166 217 166 23 44 10
2 216 413 216 44 32 12
box_min_row centroid_col centroid_row countCoords diameter_equivalent
1 12 15.08434 28.33735 166 14.53815
2 11 27.76389 25.09722 216 16.58372
euler_number filename inertia_tensor_eigenvalue1
1 0 1.jpg 73.90947
2 0 10.jpg 52.19420
inertia_tensor_eigenvalue2 label length_major_axis length_minor_axis
1 2.668406 3 34.38825 6.534103
2 20.583193 1 28.89822 18.147481
moments_hu1 moments_hu2 moments_hu3 moments_hu4 moments_hu5
1 0.4613125 0.18418092 0.0008481068 0.0005423557 3.645311e-07
2 0.3369324 0.02141752 0.0284133342 0.0015455760 -1.023864e-05
moments_hu6 moments_hu7 orientation perimeter ratio_eccentricity
1 0.0002102579 -4.918098e-08 -1.3645697 72.42031 0.9817822
2 -0.0001452330 2.732447e-07 0.3266472 92.94113 0.7782301
ratio_extent ratio_solidity
1 0.3990385 0.7649770
2 0.3214286 0.5230024trainFeatures <- trainFeatures[-1] # drop class from train; not in test
length(trainFeatures)[1] 2894length(testFeatures)[1] 2898length(intersect(testFeatures, trainFeatures)) # count of features in common[1] 2894setdiff(testFeatures, trainFeatures)[1] "Chebyshev Coefficients () [28]"
[2] "Pixel Intensity Statistics () [4]"
[3] "Radon Coefficients (Fourier ()) [2]"
[4] "Multiscale Histograms (Wavelet (Fourier ())) [0]"setdiff(trainFeatures, testFeatures)character(0)Four features must be removed from the test set. These features were constant in the training set and were removed, but they were not constant in the test set.
featuresToRemove <- setdiff(testFeatures, trainFeatures)
featuresToRemoveIndices <- which(colnames(feature.matrix) %in% featuresToRemove)
colnames(feature.matrix)[featuresToRemoveIndices][1] "Chebyshev Coefficients () [28]"
[2] "Pixel Intensity Statistics () [4]"
[3] "Radon Coefficients (Fourier ()) [2]"
[4] "Multiscale Histograms (Wavelet (Fourier ())) [0]"dim(feature.matrix)[1] 130400 2898feature.matrix <- feature.matrix[,-featuresToRemoveIndices]
dim(feature.matrix)[1] 130400 2894testFeatures <- colnames(feature.matrix)
length(trainFeatures)[1] 2894length(testFeatures)[1] 2894all(testFeatures == trainFeatures)[1] TRUEwndchrm
splits <- strsplit(row.names(feature.matrix), "-")
wndchrmFileIndex <- unlist(lapply(splits, "[", 1))
head(wndchrmFileIndex)## [1] "1" "10" "100" "1000" "10000" "100000"skimage
splits <- strsplit(skimage$filename, "\\.")
skimageFileIndex <- unlist(lapply(splits, "[", 1))
all(wndchrmFileIndex == skimageFileIndex)## [1] TRUEsum(wndchrmFileIndex == skimageFileIndex)## [1] 130400Drop columns skimage\(train and skimage\)filename since they align with corresponding info in the feature.matrix.
str(skimage)'data.frame': 130400 obs. of 30 variables:
$ area : num 166 216 1413 138 945 ...
$ area_convex : num 217 413 3673 241 1028 ...
$ area_filled : num 166 216 3478 138 945 ...
$ box_max_col : num 23 44 86 27 51 36 76 30 45 24 ...
$ box_max_row : num 44 32 72 35 51 35 48 32 43 39 ...
$ box_min_col : num 10 12 11 10 13 0 10 10 11 1 ...
$ box_min_row : num 12 11 10 10 11 12 8 10 11 3 ...
$ centroid_col : num 15.1 27.8 46.6 15.1 31.4 ...
$ centroid_row : num 28.3 25.1 40.6 22.6 30.9 ...
$ countCoords : num 166 216 1413 138 945 ...
$ diameter_equivalent : num 14.5 16.6 42.4 13.3 34.7 ...
$ euler_number : num 0 0 -17 0 0 -7 -8 -2 -1 0 ...
$ filename : chr "1.jpg" "10.jpg" "100.jpg" "1000.jpg" ...
$ inertia_tensor_eigenvalue1: num 73.9 52.2 445.2 34.4 122.1 ...
$ inertia_tensor_eigenvalue2: num 2.67 20.58 303.03 10.53 47.33 ...
$ label : num 3 1 1 1 1 1 1 1 2 1 ...
$ length_major_axis : num 34.4 28.9 84.4 23.5 44.2 ...
$ length_minor_axis : num 6.53 18.15 69.63 12.98 27.52 ...
$ moments_hu1 : num 0.461 0.337 0.53 0.326 0.179 ...
$ moments_hu2 : num 0.18418 0.02142 0.01012 0.02995 0.00625 ...
$ moments_hu3 : num 8.48e-04 2.84e-02 1.62e-04 1.37e-02 3.40e-06 ...
$ moments_hu4 : num 5.42e-04 1.55e-03 5.10e-05 3.58e-04 3.80e-07 ...
$ moments_hu5 : num 3.65e-07 -1.02e-05 4.50e-09 -4.20e-07 3.07e-13 ...
$ moments_hu6 : num 2.10e-04 -1.45e-04 1.71e-06 2.49e-05 1.42e-08 ...
$ moments_hu7 : num -4.92e-08 2.73e-07 1.12e-09 -6.70e-07 -3.05e-13 ...
$ orientation : num -1.365 0.327 0.397 -1.112 0.845 ...
$ perimeter : num 72.4 92.9 801.3 73.7 136.5 ...
$ ratio_eccentricity : num 0.982 0.778 0.565 0.833 0.782 ...
$ ratio_extent : num 0.399 0.321 0.304 0.325 0.622 ...
$ ratio_solidity : num 0.765 0.523 0.385 0.573 0.919 ...skimage$filename <- NULL # no longer needed since aligned
str(skimage)'data.frame': 130400 obs. of 29 variables:
$ area : num 166 216 1413 138 945 ...
$ area_convex : num 217 413 3673 241 1028 ...
$ area_filled : num 166 216 3478 138 945 ...
$ box_max_col : num 23 44 86 27 51 36 76 30 45 24 ...
$ box_max_row : num 44 32 72 35 51 35 48 32 43 39 ...
$ box_min_col : num 10 12 11 10 13 0 10 10 11 1 ...
$ box_min_row : num 12 11 10 10 11 12 8 10 11 3 ...
$ centroid_col : num 15.1 27.8 46.6 15.1 31.4 ...
$ centroid_row : num 28.3 25.1 40.6 22.6 30.9 ...
$ countCoords : num 166 216 1413 138 945 ...
$ diameter_equivalent : num 14.5 16.6 42.4 13.3 34.7 ...
$ euler_number : num 0 0 -17 0 0 -7 -8 -2 -1 0 ...
$ inertia_tensor_eigenvalue1: num 73.9 52.2 445.2 34.4 122.1 ...
$ inertia_tensor_eigenvalue2: num 2.67 20.58 303.03 10.53 47.33 ...
$ label : num 3 1 1 1 1 1 1 1 2 1 ...
$ length_major_axis : num 34.4 28.9 84.4 23.5 44.2 ...
$ length_minor_axis : num 6.53 18.15 69.63 12.98 27.52 ...
$ moments_hu1 : num 0.461 0.337 0.53 0.326 0.179 ...
$ moments_hu2 : num 0.18418 0.02142 0.01012 0.02995 0.00625 ...
$ moments_hu3 : num 8.48e-04 2.84e-02 1.62e-04 1.37e-02 3.40e-06 ...
$ moments_hu4 : num 5.42e-04 1.55e-03 5.10e-05 3.58e-04 3.80e-07 ...
$ moments_hu5 : num 3.65e-07 -1.02e-05 4.50e-09 -4.20e-07 3.07e-13 ...
$ moments_hu6 : num 2.10e-04 -1.45e-04 1.71e-06 2.49e-05 1.42e-08 ...
$ moments_hu7 : num -4.92e-08 2.73e-07 1.12e-09 -6.70e-07 -3.05e-13 ...
$ orientation : num -1.365 0.327 0.397 -1.112 0.845 ...
$ perimeter : num 72.4 92.9 801.3 73.7 136.5 ...
$ ratio_eccentricity : num 0.982 0.778 0.565 0.833 0.782 ...
$ ratio_extent : num 0.399 0.321 0.304 0.325 0.622 ...
$ ratio_solidity : num 0.765 0.523 0.385 0.573 0.919 ...skimage <- as.matrix(skimage)dim(feature.matrix)[1] 130400 2894dim(skimage)[1] 130400 29test.features <- cbind(feature.matrix, skimage)
rm(feature.matrix)dim(test.features)[1] 130400 2923object.size(test.features)3057898152 bytessave(test.features, file="plankton-test-wndchrm-skimage-features.Rdata")time.2 <- Sys.time()
cat(sprintf("%.1f", as.numeric(difftime(time.2, time.1, units="secs"))), " secs\n")## 224.9 secsefg @EarlGlynn
2015-02-26 0046