function input_data()
    parse_data1= Float64[]
    parse_data2 = Float64[]
    open("./dataset/data_3.txt","r") do fp
        data = readlines(fp)
        for i in 1:length(data)
            push!(parse_data1,parse(Float64,split(data[i], " ")[1]))
            push!(parse_data2 ,parse(Float64,split(data[i], " ")[2]))
        end
    end
    X = [parse_data1 parse_data2]
    return X
end

X = input_data()

using Plots
gr()

function dist(μx, μy, xx, xy)
    # ユークリッド距離
    return sqrt((xx-μx)^2+(μy-xy)^2)
end

# 重心の数
k = 2
# 重心の変数．
μx_old = Float64[]
μy_old = Float64[]
for i in 1:k
    # データの端を乱数の範囲にする．
    # 初期の重心を決める．
    push!(μx_old, rand(minimum(X[:,1]):0.1:maximum(X[:,1])))
    push!(μy_old, rand(minimum(X[:,2]):0.1:maximum(X[:,2])))
end

val_min = 1000000.0
cluster = 0
center_of_gravity_x = Float64[]
center_of_gravity_y = Float64[]
clusters = Int64[]
n = length(X[:,1])
    
μx_new = Float64[]
μy_new = Float64[]

# 収束判定用の変数
error = 100.0
error_arr = Float64[]

# プロットのgif用変数
plot_data_arr = Any[]
μx_arr = Any[]
μy_arr = Any[]


# 誤差関数が収束すれば終わり．
while error != 0.0  
    global clusters = Int64[]
    
     
    for i in 1:n
        # 重心割り当て
        # どの重心に一番近いかを決める．その距離の計算をしている．
        # iには全データが入る．
        global val_min = 100000.0
        for j in 1:k
            distance = dist(μx_old[j], μy_old[j], X[i,1], X[i,2])
            if val_min > distance
                global val_min = distance
                global cluster = j
            end
        end
        push!(clusters, cluster)
    end
    
    global plotdata = [X [string(i) for i=clusters];]
    push!(plot_data_arr, plotdata)
    
    # 主に重心の計算
    for i in 1:k
        cluster_len = length(findall((in)(["$i"]), plotdata[:,3]))
        for j in findall((in)(["$i"]), plotdata[:,3])
            # 下二行が重心の計算
            push!(center_of_gravity_x,((X[j,1]) / cluster_len))
            push!(center_of_gravity_y,((X[j,2]) / cluster_len))
        end
        push!(μx_new, sum(center_of_gravity_x))
        push!(μy_new, sum(center_of_gravity_y))
        global center_of_gravity_x = Float64[]
        global center_of_gravity_y = Float64[]
        global error = dist(μx_new[i], μy_new[i], μx_old[i], μy_old[i])
        push!(error_arr, error)
    end
    
    push!(μx_arr, μx_old)
    push!(μy_arr, μy_old)
    
    global μx_old = μx_new
    global μy_old = μy_new
    global μx_new = Float64[]
    global μy_new = Float64[]
end

tmp_x = Float64[]
tmp_y = Float64[]
x_arr = []
y_arr = []

function result(cluster_num)
    tmp_x = Float64[]
    tmp_y = Float64[]
    for p in 1:cluster_num
        result = findall((in)(["$p"]),plotdata[:,3])
        # 割り振られたデータをx座標をx_arr, y座標をy_arrにそれぞれ代入する．
        # 重心数が３だったら三つの配列で帰ってくる．
        for i in result
            push!(tmp_x, plotdata[i,1])
            push!(tmp_y, plotdata[i,2])
        end
        push!(x_arr, tmp_x)
        push!(y_arr, tmp_y)
        tmp_x = Float64[]
        tmp_y = Float64[]
    end
end

plotdata = [X [string(i) for i=clusters];]

result(k)

plot(error_arr)

scatter(x_arr, y_arr,legend=false)
scatter!(μx_old, μy_old,label="center",legend=false)

function result2(cluster_num)
    tmp_x = Float64[]
    tmp_y = Float64[]
    x_arr = []
    y_arr = []
    anim = @animate for m in 1:length(plot_data_arr)
        plotdata = plot_data_arr[m]
        μx_old = μx_arr[m]
        μy_old = μy_arr[m]
        for p in 1:cluster_num
            result = findall((in)(["$p"]), plotdata[:,3])
            # 割り振られたデータをx座標をx_arr, y座標をy_arrにそれぞれ代入する．
            # 重心数が３だったら三つの配列で帰ってくる．
            for i in result
                push!(tmp_x, plotdata[i,1])
                push!(tmp_y, plotdata[i,2])
            end
            push!(x_arr, tmp_x)
            push!(y_arr, tmp_y)
            tmp_x = Float64[]
            tmp_y = Float64[]
        end
        scatter(x_arr, y_arr,legend=false)
        scatter!(μx_old, μy_old,label="center",legend=false)
        x_arr = []
        y_arr = []
    end every 1
    gif(anim, "/Users/noriakioshita/Github/julia/ml_page/julia/kmeans.gif", fps=1)
end
result2(k)

function input_data()
    parse_data1= Float64[]
    parse_data2 = Float64[]
    open("./dataset/data_4.txt","r") do fp
        data = readlines(fp)
        for i in 1:length(data)
            push!(parse_data1,parse(Float64,split(data[i], " ")[1]))
            push!(parse_data2 ,parse(Float64,split(data[i], " ")[2]))
        end
    end
    X = [parse_data1 parse_data2]
    return X
end

X = input_data()
n = length(X[:,1])

using Plots
gr()

function dist(μx, μy, xx, xy)
    return sqrt.((xx.-μx).^2 .+(μy.-xy).^2)
end

# 重心の数
k = 9
# 重心の変数．
μx_old = Float64[]
μy_old = Float64[]

DX = Float64[]
probs = []

# プロットのgif用変数
plot_data_arr = Any[]
μx_arr = Any[]
μy_arr = Any[]


function centroid()
    # データ点からランダムに選びそれを重心とする．
    z1 = rand(1:1:length(X[:,1]))
    push!(μx_old, X[z1,1])
    push!(μy_old, X[z1,2])
    
    for i in 1:k
        DX = Float64[]
        for j in 1:n
            # 最近傍距離を計算
            Dx = minimum([dist(X[j,1], X[j,2], μx_old[m], μy_old[m]).^2 for m in 1:length(μx_old)])
            push!(DX, Dx)
        end
        probs = DX ./ sum(DX)
        cum_probs = cumsum(probs)
        r = rand()
        #println(r)
        push!(μx_old, X[findall(cum_probs -> cum_probs >= r, cum_probs)[1],1])
        push!(μy_old, X[findall(cum_probs -> cum_probs >= r, cum_probs)[1],2])
    end
end
centroid()
print(μx_old)

# 以下は全てk-meansと同じ．
min = 1000000.0
cluster = 0
center_of_gravity_x = Float64[]
center_of_gravity_y = Float64[]
clusters = Int64[]
n = length(X[:,1])
    
μx_new = Float64[]
μy_new = Float64[]

# 収束判定用の変数
error = 100.0
error_arr = Float64[]

# 誤差関数が収束すれば終わり．
while error != 0.0  
    global clusters = Int64[]
    
     
    for i in 1:n
        # 重心割り当て
        # どの重心に一番近いかを決める．その距離の計算をしている．
        # iには全データが入る．
        global min = 100000.0
        for j in 1:k
            distance = dist(μx_old[j], μy_old[j], X[i,1], X[i,2])
            if min > distance
                global min = distance
                global cluster = j
            end
        end
        push!(clusters, cluster)
    end
    
    global plotdata = [X [string(i) for i=clusters];]
    push!(plot_data_arr, plotdata)
    
    # 主に重心の計算
    for i in 1:k
        cluster_len = length(findall((in)(["$i"]), plotdata[:,3]))
        for j in findall((in)(["$i"]), plotdata[:,3])
            # 下二行が重心の計算
            push!(center_of_gravity_x,((X[j,1]) / cluster_len))
            push!(center_of_gravity_y,((X[j,2]) / cluster_len))
        end
        push!(μx_new, sum(center_of_gravity_x))
        push!(μy_new, sum(center_of_gravity_y))
        global center_of_gravity_x = Float64[]
        global center_of_gravity_y = Float64[]
        global error = dist(μx_new[i], μy_new[i], μx_old[i], μy_old[i])
        push!(error_arr, error)
    end
    
    push!(μx_arr, μx_old)
    push!(μy_arr, μy_old)
    
    global μx_old = μx_new
    global μy_old = μy_new
    global μx_new = Float64[]
    global μy_new = Float64[]
end

tmp_x = Float64[]
tmp_y = Float64[]
x_arr = []
y_arr = []

function result(cluster_num)
    tmp_x = Float64[]
    tmp_y = Float64[]
    for p in 1:cluster_num
        result = findall((in)(["$p"]),plotdata[:,3])
        # 割り振られたデータをx座標をx_arr, y座標をy_arrにそれぞれ代入する．
        # 重心数が３だったら三つの配列で帰ってくる．
        for i in result
            push!(tmp_x, plotdata[i,1])
            push!(tmp_y, plotdata[i,2])
        end
        push!(x_arr, tmp_x)
        push!(y_arr, tmp_y)
        tmp_x = Float64[]
        tmp_y = Float64[]
    end
end

plotdata = [X [string(i) for i=clusters];]

result(k)

plot(error_arr)

scatter(x_arr, y_arr,legend=false)
scatter!(μx_old, μy_old,label="center",legend=false)

function result2(cluster_num)
    tmp_x = Float64[]
    tmp_y = Float64[]
    x_arr = []
    y_arr = []
    anim = @animate for m in 1:length(plot_data_arr)
        plotdata = plot_data_arr[m]
        μx_old = μx_arr[m]
        μy_old = μy_arr[m]
        for p in 1:cluster_num
            result = findall((in)(["$p"]), plotdata[:,3])
            # 割り振られたデータをx座標をx_arr, y座標をy_arrにそれぞれ代入する．
            # 重心数が３だったら三つの配列で帰ってくる．
            for i in result
                push!(tmp_x, plotdata[i,1])
                push!(tmp_y, plotdata[i,2])
            end
            push!(x_arr, tmp_x)
            push!(y_arr, tmp_y)
            tmp_x = Float64[]
            tmp_y = Float64[]
        end
        scatter(x_arr, y_arr,legend=false)
        scatter!(μx_old, μy_old,label="center",legend=false)
        x_arr = []
        y_arr = []
    end every 1
    gif(anim, "/Users/noriakioshita/Github/julia/ml_page/julia/kmeans_plus.gif", fps=1)
    print(anim)
end
result2(k)



function input_data()
    parse_data1= Float64[]
    parse_data2 = Float64[]
    open("./dataset/data_4.txt","r") do fp
        data = readlines(fp)
        for i in 1:length(data)
            push!(parse_data1,parse(Float64,split(data[i], " ")[1]))
            push!(parse_data2 ,parse(Float64,split(data[i], " ")[2]))
        end
    end
    X = [parse_data1 parse_data2]
    return X
end

X = input_data()
n = length(X[:,1])

using Plots
gr()

function dist(μx, μy, xx, xy)
    return sqrt.((xx.-μx).^2 .+(μy.-xy).^2)
end

# 重心の数
k = 9
# 重心の変数．
μx_old = Float64[]
μy_old = Float64[]

DX = Float64[]
probs = []

# プロットのgif用変数
plot_data_arr = Any[]
μx_arr = Any[]
μy_arr = Any[]

Gram = Array{Float64}(undef, n, n)


function kernel(X)
    for i in 1:n
        for j in i:n
            if i != j
                squareDist = sqrt.((X[i,1]).^2 .+ (X[i,2]).^2)
                base = 2.0 * 4.0^2
                Gram[i, j] =exp.(-squareDist ./ base)
                Gram[j, i] = Gram[i,j]
            end
        end
    end
    return Gram
end

function centroid()
    # データ点からランダムに選びそれを重心とする．
    z1 = rand(1:1:length(X[:,1]))
    push!(μx_old, X[z1,1])
    push!(μy_old, X[z1,2])
    
    for i in 1:k
        DX = Float64[]
        for j in 1:n
            # 最近傍距離を計算
            Dx = minimum([dist(X[j,1], X[j,2], μx_old[m], μy_old[m]).^2 for m in 1:length(μx_old)])
            push!(DX, Dx)
        end
        probs = DX ./ sum(DX)
        cum_probs = cumsum(probs)
        r = rand()
        #println(r)
        push!(μx_old, X[findall(cum_probs -> cum_probs >= r, cum_probs)[1],1])
        push!(μy_old, X[findall(cum_probs -> cum_probs >= r, cum_probs)[1],2])
    end
end

X = kernel(X)
println(X)

centroid()
print(μx_old)

# 以下は全てk-meansと同じ．
val_min = 1000000.0
cluster = 0
center_of_gravity_x = Float64[]
center_of_gravity_y = Float64[]
clusters = Int64[]
n = length(X[:,1])
    
μx_new = Float64[]
μy_new = Float64[]

# 収束判定用の変数
val_error = 100.0
error_arr = Float64[]

# 誤差関数が収束すれば終わり．
while val_error != 0.0  
    global clusters = Int64[]
    
     
    for i in 1:n
        # 重心割り当て
        # どの重心に一番近いかを決める．その距離の計算をしている．
        # iには全データが入る．
        global val_min = 100000.0
        for j in 1:k
            distance = dist(μx_old[j], μy_old[j], X[i,1], X[i,2])
            if val_min > distance
                global val_min = distance
                global cluster = j
            end
        end
        push!(clusters, cluster)
    end
    
    global plotdata = [X [string(i) for i=clusters];]
    push!(plot_data_arr, plotdata)
    
    # 主に重心の計算
    for i in 1:k
        cluster_len = length(findall((in)(["$i"]), plotdata[:,3]))
        for j in findall((in)(["$i"]), plotdata[:,3])
            # 下二行が重心の計算
            push!(center_of_gravity_x,((X[j,1]) / cluster_len))
            push!(center_of_gravity_y,((X[j,2]) / cluster_len))
        end
        push!(μx_new, sum(center_of_gravity_x))
        push!(μy_new, sum(center_of_gravity_y))
        global center_of_gravity_x = Float64[]
        global center_of_gravity_y = Float64[]
        global val_error = dist(μx_new[i], μy_new[i], μx_old[i], μy_old[i])
        push!(error_arr, val_error)
    end
    
    push!(μx_arr, μx_old)
    push!(μy_arr, μy_old)
    
    global μx_old = μx_new
    global μy_old = μy_new
    global μx_new = Float64[]
    global μy_new = Float64[]
end

tmp_x = Float64[]
tmp_y = Float64[]
x_arr = []
y_arr = []

function result(cluster_num)
    tmp_x = Float64[]
    tmp_y = Float64[]
    for p in 1:cluster_num
        result = findall((in)(["$p"]),plotdata[:,3])
        # 割り振られたデータをx座標をx_arr, y座標をy_arrにそれぞれ代入する．
        # 重心数が３だったら三つの配列で帰ってくる．
        for i in result
            push!(tmp_x, plotdata[i,1])
            push!(tmp_y, plotdata[i,2])
        end
        push!(x_arr, tmp_x)
        push!(y_arr, tmp_y)
        tmp_x = Float64[]
        tmp_y = Float64[]
    end
end

plotdata = [X [string(i) for i=clusters];]

result(k)


n

for i in 1:n
    for j in i:n
        if i != j
            Gram[i,j] = 2.0
            Gram[j,i] = Gram[i,j]
        end
    end
end

Gram