交叉验证的得分怎么分析_数据探索很麻烦？推荐一款史上最强大的特征分析可视化工具：yellowbrick...

# Load the classification data set
data &＃61; load_data("occupancy")# Specify the features of interest and the classes of the target
features &＃61; ["temperature", "relative humidity", "light", "C02", "humidity"]
classes &＃61; ["unoccupied", "occupied"]# Extract the instances and target
X &＃61; data[features]
y &＃61; data.occupancy# Import the visualizer
from yellowbrick.features import RadViz# Instantiate the visualizer
visualizer &＃61; RadViz(classes&＃61;classes, features&＃61;features)visualizer.fit(X, y) # Fit the data to the visualizer
visualizer.transform(X) # Transform the data
visualizer.poof() # Draw/show/poof the data

from yellowbrick.features import Rank1D# Instantiate the 1D visualizer with the Sharpiro ranking algorithm
visualizer &＃61; Rank1D(features&＃61;features, algorithm&＃61;&＃39;shapiro&＃39;)visualizer.fit(X, y) # Fit the data to the visualizer
visualizer.transform(X) # Transform the data
visualizer.poof() # Draw/show/poof the data

# Load the classification data set
data &＃61; load_data(&＃39;concrete&＃39;)# Specify the features of interest and the target
target &＃61; "strength"
features &＃61; [&＃39;cement&＃39;, &＃39;slag&＃39;, &＃39;ash&＃39;, &＃39;water&＃39;, &＃39;splast&＃39;, &＃39;coarse&＃39;, &＃39;fine&＃39;, &＃39;age&＃39;
]# Extract the instance data and the target
X &＃61; data[features]
y &＃61; data[target]visualizer &＃61; PCADecomposition(scale&＃61;True, proj_features&＃61;True)
visualizer.fit_transform(X, y)
visualizer.poof()

import matplotlib.pyplot as pltfrom sklearn.ensemble import GradientBoostingClassifierfrom yellowbrick.features.importances import FeatureImportances# Create a new matplotlib figure
fig &＃61; plt.figure()
ax &＃61; fig.add_subplot()viz &＃61; FeatureImportances(GradientBoostingClassifier(), ax&＃61;ax)
viz.fit(X, y)
viz.poof()

from sklearn.svm import SVC
from sklearn.datasets import make_classificationfrom yellowbrick.features import RFECV# Create a dataset with only 3 informative features
X, y &＃61; make_classification(n_samples&＃61;1000, n_features&＃61;25, n_informative&＃61;3, n_redundant&＃61;2,n_repeated&＃61;0, n_classes&＃61;8, n_clusters_per_class&＃61;1, random_state&＃61;0
)# Create RFECV visualizer with linear SVM classifier
viz &＃61; RFECV(SVC(kernel&＃61;&＃39;linear&＃39;, C&＃61;1))
viz.fit(X, y)
viz.poof()

from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import StratifiedKFolddf &＃61; load_data(&＃39;credit&＃39;)target &＃61; &＃39;default&＃39;
features &＃61; [col for col in data.columns if col !&＃61; target]X &＃61; data[features]
y &＃61; data[target]cv &＃61; StratifiedKFold(5)
oz &＃61; RFECV(RandomForestClassifier(), cv&＃61;cv, scoring&＃61;&＃39;f1_weighted&＃39;)oz.fit(X, y)
oz.poof()

from sklearn.linear_model import Ridge
from yellowbrick.regressor import ResidualsPlot# Instantiate the linear model and visualizer
ridge &＃61; Ridge()
visualizer &＃61; ResidualsPlot(ridge)visualizer.fit(X_train, y_train) # Fit the training data to the model
visualizer.score(X_test, y_test) # Evaluate the model on the test data
visualizer.poof() # Draw/show/poof the data

import numpy as npfrom sklearn.linear_model import LassoCV
from yellowbrick.regressor import AlphaSelection# Create a list of alphas to cross-validate against
alphas &＃61; np.logspace(-10, 1, 400)# Instantiate the linear model and visualizer
model &＃61; LassoCV(alphas&＃61;alphas)
visualizer &＃61; AlphaSelection(model)visualizer.fit(X, y)
g &＃61; visualizer.poof()

from sklearn.ensemble import RandomForestClassifierfrom yellowbrick.classifier import ClassPredictionError# Instantiate the classification model and visualizer
visualizer &＃61; ClassPredictionError(RandomForestClassifier(), classes&＃61;classes
)# Fit the training data to the visualizer
visualizer.fit(X_train, y_train)# Evaluate the model on the test data
visualizer.score(X_test, y_test)# Draw visualization
g &＃61; visualizer.poof()


from sklearn.linear_model import LogisticRegression
from yellowbrick.classifier import DiscriminationThreshold# Instantiate the classification model and visualizer
logistic &＃61; LogisticRegression()
visualizer &＃61; DiscriminationThreshold(logistic)visualizer.fit(X, y) # Fit the training data to the visualizer
visualizer.poof() # Draw/show/poof the data


from sklearn.datasets import make_blobs# Create synthetic dataset with 8 random clusters
X, y &＃61; make_blobs(centers&＃61;8, n_features&＃61;12, shuffle&＃61;True, random_state&＃61;42)from sklearn.cluster import KMeans
from yellowbrick.cluster import KElbowVisualizer# Instantiate the clustering model and visualizer
model &＃61; KMeans()
visualizer &＃61; KElbowVisualizer(model, k&＃61;(4,12))visualizer.fit(X) # Fit the data to the visualizer
visualizer.poof() # Draw/show/poof the data


from sklearn.datasets import make_blobs# Make 12 blobs dataset
X, y &＃61; make_blobs(centers&＃61;12, n_samples&＃61;1000, n_features&＃61;16, shuffle&＃61;True)from sklearn.cluster import KMeans
from yellowbrick.cluster import InterclusterDistance# Instantiate the clustering model and visualizer
visualizer &＃61; InterclusterDistance(KMeans(9))visualizer.fit(X) # Fit the training data to the visualizer
visualizer.poof() # Draw/show/poof the data

import numpy as npfrom sklearn.tree import DecisionTreeRegressor
from yellowbrick.model_selection import ValidationCurve# Load a regression dataset
data &＃61; load_data(&＃39;energy&＃39;)# Specify features of interest and the target
targets &＃61; ["heating load", "cooling load"]
features &＃61; [col for col in data.columns if col not in targets]# Extract the instances and target
X &＃61; data[features]
y &＃61; data[targets[0]]viz &＃61; ValidationCurve(DecisionTreeRegressor(), param_name&＃61;"max_depth",param_range&＃61;np.arange(1, 11), cv&＃61;10, scoring&＃61;"r2"
)# Fit and poof the visualizer
viz.fit(X, y)
viz.poof()