fd-update

Author: tlienart

__site/A-model-tuning-hm.png

@@ -0,0 +1,9 @@
+# This file was generated, do not modify it. # hide
+using MLJ, PrettyPrinting, DataFrames, Statistics, CSV, Dates
+using PyPlot, HTTP
+MLJ.color_off() # hide
+
+req = HTTP.get("https://raw.githubusercontent.com/bbrandom91/KC_Housing/master/kc_house_data.csv")
+
+df = CSV.read(req.body, missingstring="NA")
+describe(df)

__site/A-model-tuning-hpt.svg

@@ -0,0 +1,2 @@
+# This file was generated, do not modify it. # hide
+select!(df, Not([:yr_renovated, :sqft_basement, :zipcode]));

__site/ISL-lab-5-g1.svg

@@ -0,0 +1,7 @@
+# This file was generated, do not modify it. # hide
+plt.figure(figsize=(8,6))
+plt.hist(df.price, color = "blue", edgecolor = "white", bins=50,
+         density=true)
+plt.xlabel("Price", fontsize=14)
+plt.ylabel("Frequency", fontsize=14)
+plt.savefig(joinpath(@OUTPUT, "hist_price.svg")) # hide

__site/ISL-lab-5-g3.svg

@@ -0,0 +1,10 @@
+# This file was generated, do not modify it. # hide
+plt.figure(figsize=(8,6))
+plt.hist(df.price[df.isrenovated .== true], color="blue", density=true,
+        edgecolor="white", bins=50, label="renovated", alpha=0.5)
+plt.hist(df.price[df.isrenovated .== false], color="red", density=true,
+        edgecolor="white", bins=50, label="unrenovated", alpha=0.5)
+plt.xlabel("Price", fontsize=14)
+plt.ylabel("Frequency", fontsize=14)
+plt.legend(fontsize=12)
+plt.savefig(joinpath(@OUTPUT, "hist_price2.svg")) # hide

__site/ISL-lab-5-g4.svg

@@ -0,0 +1,9 @@
+# This file was generated, do not modify it. # hide
+@load DecisionTreeRegressor
+
+y, X = unpack(df, ==(:price), col -> true)
+train, test = partition(eachindex(y), 0.7, shuffle=true, rng=5)
+
+tree = machine(DecisionTreeRegressor(), X, y)
+
+fit!(tree, rows=train);

__site/_data/kc_housing.csv

@@ -0,0 +1,2 @@
+# This file was generated, do not modify it. # hide
+rms(y[test], predict(tree, rows=test))

__site/assets/end-to-end/HouseKingCounty/code/ex1.jl

@@ -0,0 +1,2 @@
+# This file was generated, do not modify it. # hide
+@load RandomForestRegressor pkg=ScikitLearn

__site/assets/end-to-end/HouseKingCounty/code/ex10.jl

@@ -0,0 +1,2 @@
+# This file was generated, do not modify it. # hide
+coerce!(X, Finite => Count);

__site/assets/end-to-end/HouseKingCounty/code/ex11.jl

@@ -0,0 +1,6 @@
+# This file was generated, do not modify it. # hide
+rf_mdl = RandomForestRegressor()
+rf = machine(rf_mdl, X, y)
+fit!(rf, rows=train)
+
+rms(y[test], predict(rf, rows=test))

__site/assets/end-to-end/HouseKingCounty/code/ex12.jl

@@ -0,0 +1,4 @@
+# This file was generated, do not modify it. # hide
+cv3 = CV(; nfolds=3)
+res = evaluate(rf_mdl, X, y, resampling=CV(shuffle=true),
+               measure=rms, verbosity=0)

__site/assets/end-to-end/HouseKingCounty/code/ex13.jl

@@ -0,0 +1,2 @@
+# This file was generated, do not modify it. # hide
+@load XGBoostRegressor

__site/assets/end-to-end/HouseKingCounty/code/ex14.jl

@@ -0,0 +1,3 @@
+# This file was generated, do not modify it. # hide
+select!(df, Not([:id, :date]))
+schema(df)

__site/assets/end-to-end/HouseKingCounty/code/ex15.jl

@@ -0,0 +1,8 @@
+# This file was generated, do not modify it. # hide
+coerce!(X, Count => Continuous)
+
+xgb  = XGBoostRegressor()
+xgbm = machine(xgb, X, y)
+fit!(xgbm, rows=train)
+
+rms(y[test], predict(xgbm, rows=test))

__site/assets/end-to-end/HouseKingCounty/code/ex16.jl

@@ -0,0 +1,3 @@
+# This file was generated, do not modify it. # hide
+r1 = range(xgb, :max_depth, lower=3, upper=10)
+r2 = range(xgb, :num_round, lower=1, upper=25);

__site/assets/end-to-end/HouseKingCounty/code/ex17.jl

@@ -0,0 +1,8 @@
+# This file was generated, do not modify it. # hide
+tm = TunedModel(model=xgb, tuning=Grid(resolution=7),
+                resampling=CV(rng=11), ranges=[r1,r2,r3,r4,r5,r6,r7],
+                measure=rms)
+mtm = machine(tm, X, y)
+fit!(mtm, rows=train)
+
+rms(y[test], predict(mtm, rows=test))

__site/assets/end-to-end/HouseKingCounty/code/ex18.jl

@@ -0,0 +1,2 @@
+# This file was generated, do not modify it. # hide
+PyPlot.close_figs() # hide

__site/assets/end-to-end/HouseKingCounty/code/ex19.jl

@@ -0,0 +1,5 @@
+# This file was generated, do not modify it. # hide
+coerce!(df, :zipcode => Multiclass)
+df.isrenovated  = @. !ismissing(df.yr_renovated)
+df.has_basement = @. !ismissing(df.sqft_basement)
+schema(df)

__site/assets/end-to-end/HouseKingCounty/code/ex2.jl

@@ -0,0 +1,2 @@
+# This file was generated, do not modify it. # hide
+coerce!(df, :isrenovated => OrderedFactor, :has_basement => OrderedFactor);

__site/assets/end-to-end/HouseKingCounty/code/ex20.jl

@@ -0,0 +1,2 @@
+# This file was generated, do not modify it. # hide
+unique(df.waterfront)

__site/assets/end-to-end/HouseKingCounty/code/ex21.jl

@@ -0,0 +1,3 @@
+# This file was generated, do not modify it. # hide
+df.waterfront = (df.waterfront .!= "FALSE")
+coerce!(df, :waterfront => OrderedFactor);

__site/assets/end-to-end/HouseKingCounty/code/ex22.jl

@@ -0,0 +1,3 @@
+# This file was generated, do not modify it. # hide
+coerce!(df, autotype(df, :few_to_finite))
+schema(df)

__site/assets/end-to-end/HouseKingCounty/code/ex23.jl

@@ -0,0 +1,2 @@
+# This file was generated, do not modify it. # hide
+df.price = df.price ./ 1000;

__site/assets/end-to-end/HouseKingCounty/code/ex3.jl

@@ -0,0 +1,7 @@
+# This file was generated, do not modify it. # hide
+for col in names(df)
+    nmissings = sum(ismissing, df[!,col])
+    if nmissings > 0
+        println(rpad("$col has ", 25), nmissings, " missings")
+    end
+end

__site/assets/end-to-end/HouseKingCounty/code/ex4.jl

@@ -0,0 +1,25 @@
+21×8 DataFrame
+│ Row │ variable      │ mean       │ min             │ median    │ max             │ nunique │ nmissing │ eltype   │
+│     │ Symbol        │ Union…     │ Any             │ Union…    │ Any             │ Union…  │ Nothing  │ DataType │
+├─────┼───────────────┼────────────┼─────────────────┼───────────┼─────────────────┼─────────┼──────────┼──────────┤
+│ 1   │ id            │ 4.5803e9   │ 1000102         │ 3.90493e9 │ 9900000190      │         │          │ Int64    │
+│ 2   │ date          │            │ 20140502T000000 │           │ 20150527T000000 │ 372     │          │ String   │
+│ 3   │ price         │ 540088.0   │ 75000.0         │ 450000.0  │ 7.7e6           │         │          │ Float64  │
+│ 4   │ bedrooms      │ 3.37084    │ 0               │ 3.0       │ 33              │         │          │ Int64    │
+│ 5   │ bathrooms     │ 2.11476    │ 0.0             │ 2.25      │ 8.0             │         │          │ Float64  │
+│ 6   │ sqft_living   │ 2079.9     │ 290             │ 1910.0    │ 13540           │         │          │ Int64    │
+│ 7   │ sqft_lot      │ 15107.0    │ 520             │ 7618.0    │ 1651359         │         │          │ Int64    │
+│ 8   │ floors        │ 1.49431    │ 1.0             │ 1.5       │ 3.5             │         │          │ Float64  │
+│ 9   │ waterfront    │ 0.00754176 │ 0               │ 0.0       │ 1               │         │          │ Int64    │
+│ 10  │ view          │ 0.234303   │ 0               │ 0.0       │ 4               │         │          │ Int64    │
+│ 11  │ condition     │ 3.40943    │ 1               │ 3.0       │ 5               │         │          │ Int64    │
+│ 12  │ grade         │ 7.65687    │ 1               │ 7.0       │ 13              │         │          │ Int64    │
+│ 13  │ sqft_above    │ 1788.39    │ 290             │ 1560.0    │ 9410            │         │          │ Int64    │
+│ 14  │ sqft_basement │ 291.509    │ 0               │ 0.0       │ 4820            │         │          │ Int64    │
+│ 15  │ yr_built      │ 1971.01    │ 1900            │ 1975.0    │ 2015            │         │          │ Int64    │
+│ 16  │ yr_renovated  │ 84.4023    │ 0               │ 0.0       │ 2015            │         │          │ Int64    │
+│ 17  │ zipcode       │ 98077.9    │ 98001           │ 98065.0   │ 98199           │         │          │ Int64    │
+│ 18  │ lat           │ 47.5601    │ 47.1559         │ 47.5718   │ 47.7776         │         │          │ Float64  │
+│ 19  │ long          │ -122.214   │ -122.519        │ -122.23   │ -121.315        │         │          │ Float64  │
+│ 20  │ sqft_living15 │ 1986.55    │ 399             │ 1840.0    │ 6210            │         │          │ Int64    │
+│ 21  │ sqft_lot15    │ 12768.5    │ 651             │ 7620.0    │ 871200          │         │          │ Int64    │

__site/assets/end-to-end/HouseKingCounty/code/ex5.jl

@@ -0,0 +1 @@
+nothing

__site/assets/end-to-end/HouseKingCounty/code/ex6.jl

@@ -0,0 +1 @@
+nothing

__site/assets/end-to-end/HouseKingCounty/code/ex7.jl

@@ -0,0 +1 @@
+nothing

__site/assets/end-to-end/HouseKingCounty/code/ex8.jl

@@ -0,0 +1 @@
+nothing

__site/assets/end-to-end/HouseKingCounty/code/ex9.jl

@@ -0,0 +1 @@
+179.96112573828466

__site/assets/end-to-end/HouseKingCounty/code/output/ex1.out

@@ -0,0 +1,16 @@
+RandomForestRegressor(
+    n_estimators = 100,
+    criterion = "mse",
+    max_depth = nothing,
+    min_samples_split = 2,
+    min_samples_leaf = 1,
+    min_weight_fraction_leaf = 0.0,
+    max_features = "auto",
+    max_leaf_nodes = nothing,
+    min_impurity_decrease = 0.0,
+    bootstrap = true,
+    oob_score = false,
+    n_jobs = nothing,
+    random_state = nothing,
+    verbose = 0,
+    warm_start = false) @ 4…24

__site/assets/end-to-end/HouseKingCounty/code/output/ex1.res

@@ -0,0 +1 @@
+nothing

__site/assets/end-to-end/HouseKingCounty/code/output/ex10.out

@@ -0,0 +1 @@
+137.65885615237326

__site/assets/end-to-end/HouseKingCounty/code/output/ex10.res

@@ -0,0 +1,6 @@
+┌───────────┬───────────────┬────────────────────────────────────────────┐
+│ _.measure │ _.measurement │ _.per_fold                                 │
+├───────────┼───────────────┼────────────────────────────────────────────┤
+│ rms       │ 136.0         │ [148.0, 138.0, 123.0, 127.0, 143.0, 134.0] │
+└───────────┴───────────────┴────────────────────────────────────────────┘
+_.per_observation = [missing]

__site/assets/end-to-end/HouseKingCounty/code/output/ex11.out

@@ -0,0 +1,36 @@
+XGBoostRegressor(
+    num_round = 1,
+    booster = "gbtree",
+    disable_default_eval_metric = 0,
+    eta = 0.3,
+    gamma = 0.0,
+    max_depth = 6,
+    min_child_weight = 1.0,
+    max_delta_step = 0.0,
+    subsample = 1.0,
+    colsample_bytree = 1.0,
+    colsample_bylevel = 1.0,
+    lambda = 1.0,
+    alpha = 0.0,
+    tree_method = "auto",
+    sketch_eps = 0.03,
+    scale_pos_weight = 1.0,
+    updater = "auto",
+    refresh_leaf = 1,
+    process_type = "default",
+    grow_policy = "depthwise",
+    max_leaves = 0,
+    max_bin = 256,
+    predictor = "cpu_predictor",
+    sample_type = "uniform",
+    normalize_type = "tree",
+    rate_drop = 0.0,
+    one_drop = 0,
+    skip_drop = 0.0,
+    feature_selector = "cyclic",
+    top_k = 0,
+    tweedie_variance_power = 1.5,
+    objective = "reg:linear",
+    base_score = 0.5,
+    eval_metric = "rmse",
+    seed = 0) @ 9…92

__site/assets/end-to-end/HouseKingCounty/code/output/ex11.res

@@ -0,0 +1,24 @@
+┌───────────────┬─────────┬────────────┐
+│ _.names       │ _.types │ _.scitypes │
+├───────────────┼─────────┼────────────┤
+│ price         │ Float64 │ Continuous │
+│ bedrooms      │ Int64   │ Count      │
+│ bathrooms     │ Float64 │ Continuous │
+│ sqft_living   │ Int64   │ Count      │
+│ sqft_lot      │ Int64   │ Count      │
+│ floors        │ Float64 │ Continuous │
+│ waterfront    │ Int64   │ Count      │
+│ view          │ Int64   │ Count      │
+│ condition     │ Int64   │ Count      │
+│ grade         │ Int64   │ Count      │
+│ sqft_above    │ Int64   │ Count      │
+│ sqft_basement │ Int64   │ Count      │
+│ yr_built      │ Int64   │ Count      │
+│ yr_renovated  │ Int64   │ Count      │
+│ zipcode       │ Int64   │ Count      │
+│ lat           │ Float64 │ Continuous │
+│ long          │ Float64 │ Continuous │
+│ sqft_living15 │ Int64   │ Count      │
+│ sqft_lot15    │ Int64   │ Count      │
+└───────────────┴─────────┴────────────┘
+_.nrows = 21613

__site/assets/end-to-end/HouseKingCounty/code/output/ex12.out

@@ -0,0 +1 @@
+480.70583223890355

__site/assets/end-to-end/HouseKingCounty/code/output/ex12.res

@@ -0,0 +1 @@
+nothing

__site/assets/end-to-end/HouseKingCounty/code/output/ex13.out

@@ -0,0 +1 @@
+UndefVarError: r3 not defined

__site/assets/end-to-end/HouseKingCounty/code/output/ex13.res

@@ -0,0 +1 @@
+nothing

__site/assets/end-to-end/HouseKingCounty/code/output/ex14.out

@@ -0,0 +1 @@
+nothing

__site/assets/end-to-end/HouseKingCounty/code/output/ex14.res

@@ -0,0 +1,26 @@
+┌───────────────┬────────────────────────────────┬────────────────┐
+│ _.names       │ _.types                        │ _.scitypes     │
+├───────────────┼────────────────────────────────┼────────────────┤
+│ price         │ Float64                        │ Continuous     │
+│ bedrooms      │ Int64                          │ Count          │
+│ bathrooms     │ Float64                        │ Continuous     │
+│ sqft_living   │ Int64                          │ Count          │
+│ sqft_lot      │ Int64                          │ Count          │
+│ floors        │ Float64                        │ Continuous     │
+│ waterfront    │ Int64                          │ Count          │
+│ view          │ Int64                          │ Count          │
+│ condition     │ Int64                          │ Count          │
+│ grade         │ Int64                          │ Count          │
+│ sqft_above    │ Int64                          │ Count          │
+│ sqft_basement │ Int64                          │ Count          │
+│ yr_built      │ Int64                          │ Count          │
+│ yr_renovated  │ Int64                          │ Count          │
+│ zipcode       │ CategoricalValue{Int64,UInt32} │ Multiclass{70} │
+│ lat           │ Float64                        │ Continuous     │
+│ long          │ Float64                        │ Continuous     │
+│ sqft_living15 │ Int64                          │ Count          │
+│ sqft_lot15    │ Int64                          │ Count          │
+│ isrenovated   │ Bool                           │ Count          │
+│ has_basement  │ Bool                           │ Count          │
+└───────────────┴────────────────────────────────┴────────────────┘
+_.nrows = 21613

__site/assets/end-to-end/HouseKingCounty/code/output/ex15.out

@@ -0,0 +1 @@
+nothing

__site/assets/end-to-end/HouseKingCounty/code/output/ex15.res

@@ -0,0 +1,3 @@
+2-element Array{Int64,1}:
+ 0
+ 1

__site/assets/end-to-end/HouseKingCounty/code/output/ex16.out

@@ -0,0 +1 @@
+nothing

__site/assets/end-to-end/HouseKingCounty/code/output/ex16.res

@@ -0,0 +1,26 @@
+┌───────────────┬──────────────────────────────────┬───────────────────┐
+│ _.names       │ _.types                          │ _.scitypes        │
+├───────────────┼──────────────────────────────────┼───────────────────┤
+│ price         │ Float64                          │ Continuous        │
+│ bedrooms      │ CategoricalValue{Int64,UInt32}   │ OrderedFactor{13} │
+│ bathrooms     │ CategoricalValue{Float64,UInt32} │ OrderedFactor{30} │
+│ sqft_living   │ Int64                            │ Count             │
+│ sqft_lot      │ Int64                            │ Count             │
+│ floors        │ CategoricalValue{Float64,UInt32} │ OrderedFactor{6}  │
+│ waterfront    │ CategoricalValue{Bool,UInt32}    │ OrderedFactor{1}  │
+│ view          │ CategoricalValue{Int64,UInt32}   │ OrderedFactor{5}  │
+│ condition     │ CategoricalValue{Int64,UInt32}   │ OrderedFactor{5}  │
+│ grade         │ CategoricalValue{Int64,UInt32}   │ OrderedFactor{12} │
+│ sqft_above    │ Int64                            │ Count             │
+│ sqft_basement │ Int64                            │ Count             │
+│ yr_built      │ Int64                            │ Count             │
+│ yr_renovated  │ CategoricalValue{Int64,UInt32}   │ OrderedFactor{70} │
+│ zipcode       │ CategoricalValue{Int64,UInt32}   │ Multiclass{70}    │
+│ lat           │ Float64                          │ Continuous        │
+│ long          │ Float64                          │ Continuous        │
+│ sqft_living15 │ Int64                            │ Count             │
+│ sqft_lot15    │ Int64                            │ Count             │
+│ isrenovated   │ CategoricalValue{Bool,UInt32}    │ OrderedFactor{1}  │
+│ has_basement  │ CategoricalValue{Bool,UInt32}    │ OrderedFactor{1}  │
+└───────────────┴──────────────────────────────────┴───────────────────┘
+_.nrows = 21613

__site/assets/end-to-end/HouseKingCounty/code/output/ex17.out

@@ -0,0 +1 @@
+nothing

__site/assets/end-to-end/HouseKingCounty/code/output/ex17.res

@@ -0,0 +1 @@
+nothing