added beginnings of testing package (touches on good practices regard…

…ing #8)
jakewilliami · Aug 26, 2020 · 02f6e8e · 02f6e8e
1 parent 38b6e98
commit 02f6e8e
Show file tree

Hide file tree

Showing 6 changed files with 103 additions and 14 deletions.
diff --git a/Manifest.toml b/Manifest.toml
@@ -24,6 +24,12 @@ git-tree-sha1 = "fd04049c7dd78cfef0b06cdc1f0f181467655712"
 uuid = "79e6a3ab-5dfb-504d-930d-738a2a938a0e"
 version = "1.1.0"
 
+[[ArnoldiMethod]]
+deps = ["DelimitedFiles", "LinearAlgebra", "Random", "SparseArrays", "StaticArrays", "Test"]
+git-tree-sha1 = "2b6845cea546604fb4dca4e31414a6a59d39ddcd"
+uuid = "ec485272-7323-5ecc-a04f-4719b315124d"
+version = "0.0.4"
+
 [[Arpack]]
 deps = ["Arpack_jll", "Libdl", "LinearAlgebra"]
 git-tree-sha1 = "2ff92b71ba1747c5fdd541f8fc87736d82f40ec9"
@@ -104,6 +110,12 @@ git-tree-sha1 = "e58b3c2d5de62371be393f4019c27e02e7140d19"
 uuid = "059b0e18-018a-5deb-a5b2-c624ee85784b"
 version = "0.2.0"
 
+[[Clustering]]
+deps = ["Distances", "LinearAlgebra", "NearestNeighbors", "Printf", "SparseArrays", "Statistics", "StatsBase"]
+git-tree-sha1 = "b11c8d607af357776a046889a7c32567d05f1319"
+uuid = "aaaa29a8-35af-508c-8bc3-b662a17a0fe5"
+version = "0.14.1"
+
 [[CodeTracking]]
 deps = ["InteractiveUtils", "UUIDs"]
 git-tree-sha1 = "9c173f62af93cce8af2bd3527d160b6ddd6eaf81"
@@ -400,6 +412,12 @@ git-tree-sha1 = "3af30042a8fe85612a6a106cb20ca2fa1eb67bd6"
 uuid = "2996bd0c-7a13-11e9-2da2-2f5ce47296a9"
 version = "0.1.4"
 
+[[ImageSegmentation]]
+deps = ["Clustering", "DataStructures", "Distances", "ImageFiltering", "Images", "LightGraphs", "LinearAlgebra", "RegionTrees", "SimpleWeightedGraphs", "StaticArrays", "Statistics"]
+git-tree-sha1 = "3837584c12ac9170f2431c3423c2df430c386214"
+uuid = "80713f31-8817-5129-9cf8-209ff8fb23e1"
+version = "1.4.5"
+
 [[ImageShow]]
 deps = ["Base64", "FileIO", "ImageCore", "Requires"]
 git-tree-sha1 = "c9df184bc7c2e665f971079174aabb7d18f1845f"
@@ -423,6 +441,11 @@ git-tree-sha1 = "c2a145a145dc03a7620af1444e0264ef907bd44f"
 uuid = "9b13fd28-a010-5f03-acff-a1bbcff69959"
 version = "0.5.1"
 
+[[Inflate]]
+git-tree-sha1 = "f5fc07d4e706b84f72d54eedcc1c13d92fb0871c"
+uuid = "d25df0c9-e2be-5dd7-82c8-3ad0b3e990b9"
+version = "0.1.2"
+
 [[IntelOpenMP_jll]]
 deps = ["Libdl", "Pkg"]
 git-tree-sha1 = "fb8e1c7a5594ba56f9011310790e03b5384998d6"
@@ -463,9 +486,9 @@ version = "0.21.0"
 
 [[JpegTurbo_jll]]
 deps = ["Libdl", "Pkg"]
-git-tree-sha1 = "b007cedc10bb017cdae3af0062fe4dd29b483c70"
+git-tree-sha1 = "a53414ab7217ae6cc34e41c453339e17a873d169"
 uuid = "aacddb02-875f-59d6-b918-886e6ef4fbf8"
-version = "2.0.1+0"
+version = "2.0.1+1"
 
 [[Juno]]
 deps = ["Base64", "Logging", "Media", "Profile"]
@@ -492,6 +515,12 @@ git-tree-sha1 = "1fe8c3608dfe7bdec81d018de1cc66e959016e8c"
 uuid = "89763e89-9b03-5906-acba-b20f662cd828"
 version = "4.1.0+0"
 
+[[LightGraphs]]
+deps = ["ArnoldiMethod", "DataStructures", "Distributed", "Inflate", "LinearAlgebra", "Random", "SharedArrays", "SimpleTraits", "SparseArrays", "Statistics"]
+git-tree-sha1 = "6f85a35d2377cb2db1bc448ed0d6340d2bb1ea64"
+uuid = "093fc24a-ae57-5d10-9952-331d41423f4d"
+version = "1.3.3"
+
 [[LinearAlgebra]]
 deps = ["Libdl"]
 uuid = "37e2e46d-f89d-539d-b4ee-838fcccc9c8e"
@@ -552,6 +581,12 @@ git-tree-sha1 = "928b8ca9b2791081dc71a51c55347c27c618760f"
 uuid = "77ba4419-2d1f-58cd-9bb1-8ffee604a2e3"
 version = "0.3.3"
 
+[[NearestNeighbors]]
+deps = ["Distances", "StaticArrays"]
+git-tree-sha1 = "8bc6180f328f3c0ea2663935db880d34c57d6eae"
+uuid = "b8a86587-4115-5ab1-83bc-aa920d37bbce"
+version = "0.4.4"
+
 [[Netpbm]]
 deps = ["ColorTypes", "ColorVectorSpace", "FileIO", "FixedPointNumbers", "ImageCore"]
 git-tree-sha1 = "81d9a952ffd2c9146659c8427f36d98202ff4f88"
@@ -571,9 +606,9 @@ version = "1.0.4"
 
 [[OpenBLAS_jll]]
 deps = ["CompilerSupportLibraries_jll", "Libdl", "Pkg"]
-git-tree-sha1 = "1887096f6897306a4662f7c5af936da7d5d1a062"
+git-tree-sha1 = "0c922fd9634e358622e333fc58de61f05a048492"
 uuid = "4536629a-c528-5b80-bd46-f80d51c5b363"
-version = "0.3.9+4"
+version = "0.3.9+5"
 
 [[OpenSpecFun_jll]]
 deps = ["CompilerSupportLibraries_jll", "Libdl", "Pkg"]
@@ -670,6 +705,12 @@ git-tree-sha1 = "7b1d07f411bc8ddb7977ec7f377b97b158514fe0"
 uuid = "189a3867-3050-52da-a836-e630ba90ab69"
 version = "0.2.0"
 
+[[RegionTrees]]
+deps = ["IterTools", "LinearAlgebra", "StaticArrays"]
+git-tree-sha1 = "c103368232a786cc6f73480842bc471eb24c8fed"
+uuid = "dee08c22-ab7f-5625-9660-a9af2021b33f"
+version = "0.3.1"
+
 [[Requires]]
 deps = ["UUIDs"]
 git-tree-sha1 = "d37400976e98018ee840e0ca4f9d20baa231dc6b"
@@ -710,6 +751,12 @@ git-tree-sha1 = "2ee666b24ab8be6a922f9d6c11a86e1a703a7dda"
 uuid = "699a6c99-e7fa-54fc-8d76-47d257e15c1d"
 version = "0.9.2"
 
+[[SimpleWeightedGraphs]]
+deps = ["LightGraphs", "LinearAlgebra", "Markdown", "SparseArrays", "Test"]
+git-tree-sha1 = "f3f7396c2d5e9d4752357894889a87340262f904"
+uuid = "47aef6b3-ad0c-573a-a1e2-d07658019622"
+version = "1.1.1"
+
 [[Sockets]]
 uuid = "6462fe0b-24de-5631-8697-dd941f90decc"
 
@@ -821,9 +868,9 @@ version = "0.9.2"
 
 [[Zlib_jll]]
 deps = ["Libdl", "Pkg"]
-git-tree-sha1 = "a2e0d558f6031002e380a90613b199e37a8565bf"
+git-tree-sha1 = "d5bba6485811931e4b8958e2d7ca3738273ac468"
 uuid = "83775a58-1f1d-513f-b197-d71354ab007a"
-version = "1.2.11+10"
+version = "1.2.11+15"
 
 [[Zstd_jll]]
 deps = ["Libdl", "Pkg"]
@@ -845,6 +892,6 @@ version = "0.2.0"
 
 [[libpng_jll]]
 deps = ["Libdl", "Pkg", "Zlib_jll"]
-git-tree-sha1 = "594cb058723c13941cf463fd09e5859499594f50"
+git-tree-sha1 = "30518225ee32d63d034756bb5d02d782bffaf0e5"
 uuid = "b53b4c65-9356-5827-b1ea-8c7a1a84506f"
-version = "1.6.37+3"
+version = "1.6.37+4"
diff --git a/Project.toml b/Project.toml
@@ -5,6 +5,7 @@ Distributed = "8ba89e20-285c-5b6f-9357-94700520ee1b"
 FileIO = "5789e2e9-d7fb-5bc7-8068-2c6fae9b9549"
 ImageFeatures = "92ff4b2b-8094-53d3-b29d-97f740f06cef"
 ImageMagick = "6218d12a-5da1-5696-b52f-db25d2ecc6d1"
+ImageSegmentation = "80713f31-8817-5129-9cf8-209ff8fb23e1"
 ImageTransformations = "02fcd773-0e25-5acc-982a-7f6622650795"
 Images = "916415d5-f1e6-5110-898d-aaa5f9f070e0"
 Netpbm = "f09324ee-3d7c-5217-9330-fc30815ba969"

diff --git a/src/Adaboost.jl b/src/Adaboost.jl
@@ -9,7 +9,7 @@ using Distributed # for parallel processing (namely, @everywhere)
 
 include("HaarLikeFeature.jl")
 
-
+import Base.print_matrix
 
 # TODO: select optimal threshold for each feature
 # TODO: attentional cascading
@@ -49,6 +49,7 @@ function learn(positiveIIs::AbstractArray, negativeIIs::AbstractArray, numClassi
     # Initialise weights $w_{1,i} = \frac{1}{2m}, \frac{1}{2l}$, for $y_i=0,1$ for negative and positive examples respectively
     posWeights = deepfloat(ones(numPos)) / (2 * numPos)
     negWeights = deepfloat(ones(numNeg)) / (2 * numNeg)
+    # println(negWeights)
     #=
     Consider the original code,
         ```
@@ -65,19 +66,23 @@ function learn(positiveIIs::AbstractArray, negativeIIs::AbstractArray, numClassi
     # Concatenate positive and negative weights into one `weights` array
     weights = vcat(posWeights, negWeights)
     # println(weights)
+    # print_matrix(stdout, weights)
     # weights = vcat(posWeights, negWeights)
     # weights = hcat((posWeights, negWeights))
     # weights = vcat([posWeights, negWeights])
     # labels = vcat((ones(numPos), ones(numNeg) * -1))
     labels = vcat(ones(numPos), ones(numNeg) * -1)
+    # println(labels)
     # labels = vcat([ones(numPos), ones(numNeg) * -1])
 
     # get list of images (global because of @everywhere scope)
     # global images = positiveIIs + negativeIIs
     global images = vcat(positiveIIs, negativeIIs)
+    # println(images)
 
     # Create features for all sizes and locations
     global features = _create_features(imgHeight, imgWidth, minFeatureWidth, maxFeatureWidth, minFeatureHeight, maxFeatureHeight)
+    # [println(f.weight) for f in features]
     numFeatures = length(features)
     # feature_indexes = list(range(num_features))
     featureIndices = Array(1:numFeatures)
@@ -116,8 +121,11 @@ function learn(positiveIIs::AbstractArray, negativeIIs::AbstractArray, numClassi
 
     n = numClassifiers
     p = Progress(n, 1)   # minimum update interval: 1 second
-    for t in processes
+    for t in numClassifiers
+    # for t in processes
         # println(typeof(length(featureIndices)))
+        # print_matrix(stdout, weights)
+        println(weights)
         classificationErrors = zeros(length(featureIndices))
 
         # normalize the weights $w_{t,i}\gets \frac{w_{t,i}}{\sum_{j=1}^n w_{t,j}}$
@@ -127,7 +135,11 @@ function learn(positiveIIs::AbstractArray, negativeIIs::AbstractArray, numClassi
         # weights *= 1. / sum(weights)
         # weights = weights/sum(weights)
         # weights *= 1. / sum(weights)
+        # println(weights)
+        # print_matrix(stdout, weights)
         weights = deepdiv(deepfloat(weights), deepsum(weights))
+        # println(weights)
+        # print_matrix(stdout, weights)
 
         # For each feature j, train a classifier $h_j$ which is restricted to using a single feature.  The error is evaluated with respect to $w_j,\varepsilon_j = \sum_i w_i\left|h_j\left(x_i\right)-y_i\right|$
         for j in 1:length(featureIndices)
@@ -137,6 +149,8 @@ function learn(positiveIIs::AbstractArray, negativeIIs::AbstractArray, numClassi
             # error = sum(imgIDX -> (labels[imgIDX] ≠ votes[imgIDX, fIDX]) ? weights[imgIDX] : 0, 1:numImgs)
             # error = sum(map(lambda img_idx: weights[img_idx] if labels[img_idx] != votes[img_idx, f_idx] else 0, range(num_imgs)))
             ε = deepsum(map(imgIDX -> labels[imgIDX] ≠ votes[imgIDX, fIDX] ? weights[imgIDX] : 0, 1:numImgs))
+            # println(weights)
+            # map(imgIDX -> labels[imgIDX] ≠ votes[imgIDX, fIDX] ? println(weights[imgIDX]) : println(0), 1:numImgs)
             # lambda img_idx: weights[img_idx] if labels[img_idx] != votes[img_idx, f_idx] else 0
             # imgIDX -> (labels[imgIDX] ≠ votes[imgIDX, fIDX]) ? weights[imgIDX] : 0
 
@@ -152,9 +166,10 @@ function learn(positiveIIs::AbstractArray, negativeIIs::AbstractArray, numClassi
         # println(weights)
         # set feature weight
         bestFeature = features[bestFeatureIDX]
-        featureWeight = 0.5 * log((1 - bestError) / bestError)
+        featureWeight = 0.5 * log((1 - bestError) / bestError) # β
         # featureWeight = (1 - bestError) / bestError # β
         # println(typeof(featureWeight))
+        # [println(f.weight) for f in features]
         bestFeature.weight = featureWeight
 
         # classifiers = vcat(classifiers, bestFeature)
@@ -181,6 +196,7 @@ function learn(positiveIIs::AbstractArray, negativeIIs::AbstractArray, numClassi
         # remove feature (a feature can't be selected twice)
         # feature_indexes.remove(best_feature_idx)
         featureIndices = filter!(e -> e ∉ bestFeatureIDX, featureIndices) # note: without unicode operators, `e ∉ [a, b]` is `!(e in [a, b])`
+        # println(bestFeature)
 
         next!(p)
     end
@@ -217,6 +233,7 @@ function _create_features(imgHeight::Int64, imgWidth::Int64, minFeatureWidth::In
     # features = Array()
     features = []
 
+
     # for feature in FeatureTypes # from HaarLikeFeature.jl
     #     # FeatureTypes are just tuples
     #     println(typeof(feature), " " , feature)
@@ -233,8 +250,11 @@ function _create_features(imgHeight::Int64, imgWidth::Int64, minFeatureWidth::In
                     for y in 1:(imgHeight - featureHeight)
                         # println("top left: ", typeof((x, y)), " ", (x, y))
                         features = push!(features, HaarLikeFeature(feature, (x, y), featureWidth, featureHeight, 0, 1))
+
                         # features = (features..., HaarLikeFeature(feature, (x, y), featureWidth, featureHeight, 0, 1)) # using splatting to add to a tuple.  see Utils.partial()
                         features = push!(features, HaarLikeFeature(feature, (x, y), featureWidth, featureHeight, 0, -1))
+                        # [println(f.weight) for f in features]
+                        # println(HaarLikeFeature(feature, (x, y), featureWidth, featureHeight, 0, 1).weight)
                         # features = (features..., HaarLikeFeature(feature, (x, y), featureWidth, featureHeight, 0, -1))
                         # feature.append(HaarLikeFeature...)
                     end # end for y
@@ -244,7 +264,7 @@ function _create_features(imgHeight::Int64, imgWidth::Int64, minFeatureWidth::In
     end # end for feature in feature types
 
     # println("...finished processing; ", length(features), " features created.")
-
+    # [println(f.weight) for f in features]
     return features
 end
 

diff --git a/src/FDA.jl b/src/FDA.jl
@@ -41,6 +41,7 @@ function main(alt::Bool=false)
       maxFeatureWidth = 10
 
 
+      ### TRAINING PROCESS
       println("Loading faces...")
 
       facesTraining = loadImages(posTrainingPath)
@@ -96,7 +97,16 @@ function main(alt::Bool=false)
       # recon.save("/Users/jakeireland/Desktop/reconstruction.png")
 
       # println(reconstructedImage)
-      save("/Users/jakeireland/Desktop/reconstruction.png", colorview(Gray, reconstructedImage))
+      # save("/Users/jakeireland/Desktop/reconstruction.png", colorview(Gray, map(clamp01nan, reconstructedImage)))
+      # save("/Users/jakeireland/Desktop/reconstruction.png", reconstructedImage)
+      # readblob(reconstructedImage, "/Users/jakeireland/Desktop/reconstruction.png")
+
+      # save("/Users/jakeireland/Desktop/reconstruction.png", reconstructedImage)
+      save("/Users/jakeireland/Desktop/reconstruction.png", Gray.(map(clamp01nan, reconstructedImage)))
+      # save("/Users/jakeireland/Desktop/reconstruction.png", Gray.(reconstructedImage ./ 255))
+      # println(channelview(reconstructedImage))
+      # colorview(Gray, reconstructedImage)
+
 end
 
 

diff --git a/src/HaarLikeFeature.jl b/src/HaarLikeFeature.jl
@@ -44,8 +44,9 @@ mutable struct HaarLikeFeature <: HaarFeatureType#struct HaarLikeFeature{T} <: H
         topLeft = position
         bottomRight = (position[1] + width, position[2] + height)
         weight = 1
+        # println(weight)
 
-        new(featureType, position, topLeft, bottomRight, width, height, threshold, polarity)
+        new(featureType, position, topLeft, bottomRight, width, height, threshold, polarity, weight)
     end # end constructor
 end # end structure
 

diff --git a/test/runtests.jl b/test/runtests.jl
@@ -0,0 +1,10 @@
+#!/usr/bin/env bash
+    #=
+    exec julia --project="~/FaceDetection.jl/" "${BASH_SOURCE[0]}" "$@" -e "include(popfirst!(ARGS))" \
+    "${BASH_SOURCE[0]}" "$@"
+    =#
+
+using Test
+using FaceDetection
+
+# @test ... == ...