update

L10/Extra/demo.py

+#!/opt/anaconda3/bin/python3
+
+import numpy as np
+import matplotlib
+import matplotlib.pyplot as plt
+
+def LoadDataFromL01():
+    import pickle
+    filename = "../../L04/Data/itmal_l01_data.pkl"
+    with open(f"{filename}", "rb") as f:
+        (X, y) = pickle.load(f)
+        return X, y
+
+X, y = LoadDataFromL01()
+
+print(f"X.shape={X.shape},  y.shape={y.shape}")
+
+assert X.shape[0] == y.shape[0]
+assert X.ndim == 2
+assert y.ndim == 1  # did a y.ravel() before saving to picke file
+assert X.shape[0] == 29
+
+# re-create plot data (not stored in the Pickel file)
+m = np.linspace(0, 60000, 1000)
+M = np.empty([m.shape[0], 1])
+M[:, 0] = m
+
+print("OK")
+
+
+# Setup the MLP and lin. regression again..
+
+from sklearn.neural_network import MLPRegressor
+from sklearn.linear_model import LinearRegression
+from sklearn.model_selection import cross_val_predict
+from sklearn.metrics import r2_score
+
+def PlotModels(model1, model2, X, y, name_model1="lin.reg", name_model2="MLP"):
+
+    # NOTE: local function is such a nifty feature of Python!
+    def CalcPredAndScore(model1, model2, X, y):
+        y_pred_model1 = model1.predict(X)
+        y_pred_model2 = model2.predict(X)
+
+        # call r2
+        score_model1 = r2_score(y, y_pred_model1)
+        score_model2 = r2_score(y, y_pred_model2)
+
+        return y_pred_model1, y_pred_model2, score_model1, score_model2
+
+    def Fill(s, n):
+        while(len(s)<n):
+            s += " "
+        return s
+
+    y_pred_model1, y_pred_model2, score_model1, score_model2 = CalcPredAndScore(
+        model1, model2, X, y)
+
+    plt.plot(X, y_pred_model1, "r.-")
+    plt.plot(X, y_pred_model2, "kx-")
+    plt.scatter(X, y)
+    plt.xlabel("GDP per capita")
+    plt.ylabel("Life satisfaction")
+    plt.legend([name_model1, name_model2, "X OECD data"])
+
+    l = max(len(name_model1), len(name_model2))
+
+    print(f"{Fill(name_model1,l)}.score(X, y)={score_model1:0.2f}")
+    print(f"{Fill(name_model2,l)}.score(X, y)={score_model2:0.2f}")
+
+
+# lets make a linear and MLP regressor and redo the plots
+
+mlp = MLPRegressor(hidden_layer_sizes=(10, ),
+                   solver='lbfgs',
+                   activation='relu',
+                   tol=1E-5,
+                   max_iter=100000,
+                   verbose=False)
+linreg = LinearRegression()
+
+mlp.fit(X, y)
+linreg.fit(X, y)
+
+print("The MLP may mis-fit the data, seen in the, sometimes, bad R^2 score..\n")
+PlotModels(linreg, mlp, X, y)
+
+print("\nOK")
+
+
+from sklearn.neural_network import MLPRegressor
+solvers = {"lbfgs", "sgd", "adam"}
+
+mlps = {}
+y_preds = {}
+for i in solvers:
+    mlp = MLPRegressor(solver=i, hidden_layer_sizes=(1,))
+
+    mlp.fit(X, y.ravel())
+    score  = mlp.score(X, y)
+
+    y_preds[i] = mlp.predict(M)
+    mlps[i] = mlp
+
+    print(f"mlp(solver={i:5},..)  => score={score:.1f}")
+
+print()
+PlotModels(linreg, mlps["lbfgs"], X, y, "lin.reg", "MPL[lbfgs]")
+
+print("\nOK")