Related
I am using XGboost for a binary prediction problem. I tested my model with several features and had some good results.
After adding one feature to the model and calculating the feature importance. The importance of this feature showed to be very high and far superior to other features.
However, when testing the model the test score drops considerably.
Is there an explanation for this kind of behaviour ?
There are at least a few ways to run feature importance experiments.
# Let's load the packages
import numpy as np
import pandas as pd
from sklearn.datasets import load_boston
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestRegressor
from sklearn.inspection import permutation_importance
import shap
from matplotlib import pyplot as plt
plt.rcParams.update({'figure.figsize': (12.0, 8.0)})
plt.rcParams.update({'font.size': 14})
boston = load_boston()
X = pd.DataFrame(boston.data, columns=boston.feature_names)
y = boston.target
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=12)
rf = RandomForestRegressor(n_estimators=100)
rf.fit(X_train, y_train)
# 1
rf.feature_importances_
plt.barh(boston.feature_names, rf.feature_importances_)
sorted_idx = rf.feature_importances_.argsort()
plt.barh(boston.feature_names[sorted_idx], rf.feature_importances_[sorted_idx])
plt.xlabel("Random Forest Feature Importance")
# 2
perm_importance = permutation_importance(rf, X_test, y_test)
sorted_idx = perm_importance.importances_mean.argsort()
plt.barh(boston.feature_names[sorted_idx], perm_importance.importances_mean[sorted_idx])
plt.xlabel("Permutation Importance")
# 3
explainer = shap.TreeExplainer(rf)
shap_values = explainer.shap_values(X_test)
shap.summary_plot(shap_values, X_test, plot_type="bar")
Also, you can certainly add more data into your model. Models, almost without exception, produce more accurate results when they 'see' more data. Finally, you can always test other models on your dataset and see how they perform. Today at work I tested an XGboost model and a RandomForestRegressor model. I expected the former to perform better, but the latter actually performed much better. It's almost impossible to guess which model will perform better over any given dataset, you have to try multiple models, check the predictive capabilities of each, and pick the one (or maybe two) that performs the best. Having said that, you can try something like this.
import time
import numpy as np
import matplotlib.pyplot as plt
from sklearn import cluster, datasets
from sklearn.neighbors import kneighbors_graph
from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import train_test_split
np.random.seed(0)
pd.set_option('display.max_columns', 500)
#df = pd.read_csv('C:\\your_path_here\\test.csv')
#print('done!')
#df = df[:10000]
#df = df.fillna(0)
#df = df.dropna()
X = df[['RatingScore',
'Par',
'Term',
'TimeToMaturity',
'LRMScore',
'Coupon',
'Price']]
#select your target variable
y = df[['Spread']]
#train test split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, random_state=42)
colors = np.array([x for x in 'bgrcmykbgrcmykbgrcmykbgrcmyk'])
colors = np.hstack([colors] * 20)
clustering_names = [
'MiniBatchKMeans', 'AffinityPropagation', 'MeanShift',
'SpectralClustering', 'Ward', 'AgglomerativeClustering',
'DBSCAN', 'Birch']
plt.figure(figsize=(len(clustering_names) * 2 + 3, 9.5))
plt.subplots_adjust(left=.02, right=.98, bottom=.001, top=.96, wspace=.05,
hspace=.01)
plot_num = 1
blobs = datasets.make_blobs(n_samples=n_samples, random_state=8)
# normalize dataset for easier parameter selection
X = StandardScaler().fit_transform(X)
# estimate bandwidth for mean shift
bandwidth = cluster.estimate_bandwidth(X, quantile=0.3)
# connectivity matrix for structured Ward
connectivity = kneighbors_graph(X, n_neighbors=10, include_self=False)
# make connectivity symmetric
connectivity = 0.5 * (connectivity + connectivity.T)
# create clustering estimators
ms = cluster.MeanShift(bandwidth=bandwidth, bin_seeding=True)
two_means = cluster.MiniBatchKMeans(n_clusters=2)
ward = cluster.AgglomerativeClustering(n_clusters=2, linkage='ward',
connectivity=connectivity)
spectral = cluster.SpectralClustering(n_clusters=2,
eigen_solver='arpack',
affinity="nearest_neighbors")
dbscan = cluster.DBSCAN(eps=.2)
affinity_propagation = cluster.AffinityPropagation(damping=.9,
preference=-200)
average_linkage = cluster.AgglomerativeClustering(
linkage="average", affinity="cityblock", n_clusters=2,
connectivity=connectivity)
birch = cluster.Birch(n_clusters=2)
clustering_algorithms = [
two_means, affinity_propagation, ms, spectral, ward, average_linkage,
dbscan, birch]
for name, algorithm in zip(clustering_names, clustering_algorithms):
# predict cluster memberships
t0 = time.time()
algorithm.fit(X)
t1 = time.time()
if hasattr(algorithm, 'labels_'):
y_pred = algorithm.labels_.astype(np.int)
else:
y_pred = algorithm.predict(X)
# plot
plt.subplot(4, len(clustering_algorithms), plot_num)
if i_dataset == 0:
plt.title(name, size=18)
plt.scatter(X[:, 0], X[:, 1], color=colors[y_pred].tolist(), s=10)
if hasattr(algorithm, 'cluster_centers_'):
centers = algorithm.cluster_centers_
center_colors = colors[:len(centers)]
plt.scatter(centers[:, 0], centers[:, 1], s=100, c=center_colors)
plt.xlim(-2, 2)
plt.ylim(-2, 2)
plt.xticks(())
plt.yticks(())
plt.text(.99, .01, ('%.2fs' % (t1 - t0)).lstrip('0'),
transform=plt.gca().transAxes, size=15,
horizontalalignment='right')
plot_num += 1
plt.show()
Finally, consider looping through several regression, or classification, models in one go, and getting the results for each.
from sklearn.linear_model import LinearRegression
regressor = LinearRegression()
regressor.fit(X_train, y_train)
from sklearn import linear_model
import statsmodels.api as sm
X = X
y = y
# Note the difference in argument order
model = sm.OLS(y, X).fit()
predictions = model.predict(X) # make the predictions by the model
# Print out the statistics
model.summary()
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 0)
from sklearn.linear_model import LinearRegression
from sklearn.linear_model import SGDRegressor
from sklearn.neighbors import KNeighborsRegressor
from sklearn.tree import DecisionTreeRegressor
from sklearn.ensemble import RandomForestRegressor
from sklearn.ensemble import GradientBoostingRegressor
from sklearn.linear_model import TweedieRegressor
from sklearn.linear_model import PoissonRegressor
from sklearn.linear_model import Ridge
from sklearn.linear_model import Lasso
from sklearn.svm import LinearSVR
from sklearn.metrics import mean_squared_error
from sklearn.metrics import mean_absolute_error
regressors = [
LinearRegression(),
SGDRegressor(),
KNeighborsRegressor(),
DecisionTreeRegressor(),
RandomForestRegressor(),
GradientBoostingRegressor(),
TweedieRegressor(),
PoissonRegressor(),
Ridge(),
Lasso()
]
import pandas as pd
# Logging for Visual Comparison
log_cols=["Regressor", "RMSE", "MAE"]
log = pd.DataFrame(columns=log_cols)
for reg in regressors:
reg.fit(X_train, y_train)
name = reg.__class__.__name__
print(reg.score(X_test, y_test))
y_pred = reg.predict(X_test)
lr_mse = mean_squared_error(y_pred, y_test)
lr_rmse = np.sqrt(lr_mse)
print(name + ' RMSE: %.4f' % lr_rmse)
lin_mae = mean_absolute_error(y_pred, y_test)
print(name + ' MAE: %.4f' % lin_mae)
log_entry = pd.DataFrame([[name, lr_rmse, lin_mae]], columns=log_cols)
log = log.append(log_entry)
print("="*30)
import seaborn as sns
import matplotlib as plt
sns.set_color_codes("muted")
sns.barplot(x='RMSE', y='Regressor', data=log, color="b")
import numpy as np
import matplotlib.pyplot as plt
from matplotlib.colors import ListedColormap
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.datasets import make_moons, make_circles, make_classification
from sklearn.neural_network import MLPClassifier
from sklearn.gaussian_process.kernels import RBF
from sklearn.naive_bayes import GaussianNB
from sklearn.metrics import accuracy_score, log_loss
from sklearn.neighbors import KNeighborsClassifier
from sklearn.gaussian_process import GaussianProcessClassifier
from sklearn.tree import DecisionTreeClassifier
from sklearn.ensemble import RandomForestClassifier, AdaBoostClassifier, GradientBoostingClassifier
from sklearn.discriminant_analysis import QuadraticDiscriminantAnalysis
from sklearn.svm import SVC
from sklearn.datasets import load_iris
iris = load_iris()
iris
# Step 2: Separating the data into dependent and independent variables
X = iris.data[:, :2] # we only take the first two features.
y = iris.target
# Step 3: Splitting the dataset into the Training set and Test set
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 0)
classifiers = [
GaussianNB(),
MLPClassifier(),
KNeighborsClassifier(),
GaussianProcessClassifier(),
DecisionTreeClassifier(),
RandomForestClassifier(),
AdaBoostClassifier(),
GradientBoostingClassifier(),
QuadraticDiscriminantAnalysis()]
import pandas as pd
# Logging for Visual Comparison
log_cols=["Classifier", "Accuracy"]
log = pd.DataFrame(columns=log_cols)
for clf in classifiers:
clf.fit(X_train, y_train)
name = clf.__class__.__name__
print("="*30)
print(name)
print('****Results****')
train_predictions = clf.predict(X_test)
acc = accuracy_score(y_test, train_predictions)
print("Accuracy: {:.4%}".format(acc))
log_entry = pd.DataFrame([[name, acc*100]], columns=log_cols)
log = log.append(log_entry)
print("="*30)
import seaborn as sns
import matplotlib as plt
sns.set_color_codes("muted")
sns.barplot(x='Accuracy', y='Classifier', data=log, color="b")
I am trying to get SHAP values for a Gaussian Processes Regression (GPR) model using SHAP library. However, all SHAP values are zero. I am using the example in the official documentation. I only changed the model to GPR.
import sklearn
from sklearn.model_selection import train_test_split
import numpy as np
import shap
import time
from sklearn.gaussian_process import GaussianProcessRegressor
from sklearn.gaussian_process.kernels import Matern, WhiteKernel, ConstantKernel
shap.initjs()
X,y = shap.datasets.diabetes()
X_train,X_test,y_train,y_test = train_test_split(X, y, test_size=0.2, random_state=0)
# rather than use the whole training set to estimate expected values, we summarize with
# a set of weighted kmeans, each weighted by the number of points they represent.
X_train_summary = shap.kmeans(X_train, 10)
kernel = Matern(length_scale=2, nu=3/2) + WhiteKernel(noise_level=1)
gp = GaussianProcessRegressor(kernel)
gp.fit(X_train, y_train)
# explain all the predictions in the test set
explainer = shap.KernelExplainer(gp.predict, X_train_summary)
shap_values = explainer.shap_values(X_test)
shap.summary_plot(shap_values, X_test)
Running the above code gives the following plot:
When I use Neural Network or Linear Regression, the above code works fine without problem.
If you have any idea how to solve this issue, please let me know.
Your model doesn't predict anything:
plt.scatter(y_test, gp.predict(X_test));
Train your model properly, like below:
plt.scatter(y_test, gp.predict(X_test));
and you're fine to go:
explainer = shap.KernelExplainer(gp.predict, X_train_summary)
shap_values = explainer.shap_values(X_test)
shap.summary_plot(shap_values, X_test)
Full reproducible example:
import sklearn
from sklearn.model_selection import train_test_split
import numpy as np
import shap
import time
from sklearn.gaussian_process import GaussianProcessRegressor
from sklearn.gaussian_process.kernels import WhiteKernel, DotProduct
X,y = shap.datasets.diabetes()
X_train,X_test,y_train,y_test = train_test_split(X, y, test_size=0.2, random_state=0)
X_train_summary = shap.kmeans(X_train, 10)
kernel = DotProduct() + WhiteKernel()
gp = GaussianProcessRegressor(kernel)
gp.fit(X_train, y_train)
explainer = shap.KernelExplainer(gp.predict, X_train_summary)
shap_values = explainer.shap_values(X_test)
shap.summary_plot(shap_values, X_test)
Try this code:
kernel = 1.0 * Matern(length_scale=1.0, nu=2.5) + \
WhiteKernel(noise_level=10**-1,noise_level_bounds=(10**-1, 10**1))
model = GaussianProcessRegressor(kernel=kernel,
optimizer='fmin_l_bfgs_b',random_state=123)
explainer = shap.Explainer(model.predict,X_train)
shap_values = explainer.shap_values(X_train)
shap.plots.bar(shap_values) ## bar plot
shap.summary_plot(shap_values, X_train,show=False) ## summary
I am performing gridsearchcv on ridgeclassifiercv to obtain hyper-parameters for my model.
So i imported the libraries
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.model_selection import train_test_split, cross_val_score
from sklearn.preprocessing import StandardScaler
from sklearn import linear_model
from sklearn.model_selection import GridSearchCV, RandomizedSearchCV
import warnings
warnings.filterwarnings('ignore')
np.random.seed(27)
Then i imported the dataset and split, scaled and label encoded the target variable
!wget -O ChurnData.csv https://s3-api.us-geo.objectstorage.softlayer.net/cf-courses-data/CognitiveClass/ML0101ENv3/labs/ChurnData.csv
churn = pd.read_csv("ChurnData.csv")
X = churn.drop(['churn'], axis='columns')
y1 = churn[['churn']]
y1['churn']=y1['churn'].astype('int')
scaler=StandardScaler()
X_scaled=scaler.fit_transform(X)
from sklearn import preprocessing
le = preprocessing.LabelEncoder()
le.fit(churn['churn'].unique())
y = le.transform(y1)
# split
X_train, X_test, y_train, y_test = train_test_split(X_scaled, y, test_size = 0.2)
Then i performed gridsearchcv
alphas = [(0.1, 1, 2, 5 , 10)]
solver_churn = ['auto', 'svd','cholesky', 'lsqr', 'sparse_cg', 'sag', 'saga']
fit_intercept = [True, False]
class_weight = [{0:0.5,1:0.5},{0:0.6,1:0.4}]
param_grid_churn = dict(alphas=alphas, fit_intercept=fit_intercept,class_weight=class_weight)
ridgecv = linear_model.RidgeClassifierCV()
grids_churn = GridSearchCV(estimator=ridgecv, param_grid=param_grid_churn, scoring='roc_auc', verbose=1, n_jobs=-1)
grid_result_churn = grids_churn.fit(X_train, y_train)
alphas is given in docs as a parameter still i get
Error in Grid search CV - RidgeClassifierCV as the constructor either does not set or modifies parameter alphas
How to resolve this?
Adjust your code like this:
alphas = (0.1, 1, 2, 5 , 10)
solver_churn = ['auto', 'svd','cholesky', 'lsqr', 'sparse_cg', 'sag', 'saga']
fit_intercept = [True, False]
class_weight = [{0:0.5,1:0.5},{0:0.6,1:0.4}]
param_grid_churn = dict(fit_intercept=fit_intercept,class_weight=class_weight)
ridgecv = linear_model.RidgeClassifierCV(alphas=alphas)
grids_churn = GridSearchCV(estimator=ridgecv, param_grid=param_grid_churn, scoring='roc_auc', verbose=1, n_jobs=-1)
grid_result_churn = grids_churn.fit(X_train, y_train)
import math
import numpy as np
import pandas as pd
#from pandas import DataFrame
from sklearn import preprocessing,cross_validation
from sklearn.linear_model import LogisticRegression
#from sklearn.cross_validation import train_test_split
from numpy import loadtxt, where
from pylab import scatter, show, legend, xlabel, ylabel
# scale larger positive and values to between -1,1 depending on the largest
# value in the data
min_max_scaler = preprocessing.MinMaxScaler(feature_range=(0, 1))
df = pd.read_excel("Cryotherapy.xlsx", header=0)
# clean up data
df.columns = ["sex","age","Time","Number_of_Warts", "Type",
"Area","Result_of_Treatment"]
x = df["Result_of_Treatment"]
X = df[["Type","Area",]]
X = np.array(X)
X = min_max_scaler.fit_transform(X)
Y = df["Result_of_Treatment"]
Y = np.array(Y)
X_train, X_test, Y_train, Y_test = cross_validation.train_test_split(X, Y,
test_size=0.4)
# train scikit learn model
clf = LogisticRegression()
clf.fit(X_train, Y_train)
accuracy = clf.score(X_test,Y_test)
print(accuracy)
Try passing a random_state into the train_test_split function. If you do not do this then the data is gonna be shuffled randomly each time -> producing different train and test sets.
Example:
X_train, X_test, Y_train, Y_test = cross_validation.train_test_split(X, Y, test_size=0.4, random_state=1)
I am new to python and also learning machine learning. I got a data-set for titanic and trying to predict who survived and who did not. But my code seems to have an issue with the y_pred, as none of them is close to 1 or above one. Find attached also the y_test and y_pred images.
# Importing the libraries
import numpy as np
import matplotlib.pyplot as plt
import pandas as pd
# Importing the dataset
dataset = pd.read_csv('train.csv')
X = dataset.iloc[:, :-1].values
y = dataset.iloc[:, 3].values
# Taking care of missing data
from sklearn.preprocessing import Imputer
imputer = Imputer(missing_values = 'NaN', strategy = 'mean', axis = 0)
imputer = imputer.fit(X[:, 2:3])
X[:, 2:3] = imputer.transform(X[:, 2:3])
#Encoding Categorical variable
from sklearn.preprocessing import LabelEncoder, OneHotEncoder
labelencoder_X = LabelEncoder()
X[:, 0] = labelencoder_X.fit_transform(X[:, 0])
onehotencoder = OneHotEncoder(categorical_features = [0])
X = onehotencoder.fit_transform(X).toarray()
# Dummy variable trap
X = X[:, 1:]
# Splitting the Dataset into Training Set and Test Set
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 0)
# Split the dataset into training and test set
from sklearn.model_selection import train_test_split
X_train, X_test, y_tratin, y_test = train_test_split(X, y, test_size = 0.2,)
# Fitting the Multiple Linear Regression to the training set
""" regressor is an object of LinearRegression() class in line 36 """
from sklearn.linear_model import LinearRegression
regressor = LinearRegression()
regressor.fit(X_train, y_train)
Thanks for the help everyone, I have been able to sort it out.
The problem was y in the importing dataset was seen as a vector and not a matrix
# Importing the libraries
import numpy as np
import matplotlib.pyplot as plt
import pandas as pd
# Importing the dataset
dataset = pd.read_csv('train.csv')
X = dataset.iloc[:, :-1].values
y = dataset.iloc[:, 3:].values
# Taking care of missing data
from sklearn.preprocessing import Imputer
imputer = Imputer(missing_values = 'NaN', strategy = 'mean', axis = 0)
imputer = imputer.fit(X[:, 2:3])
X[:, 2:3] = imputer.transform(X[:, 2:3])
#Encoding Categorical variable
from sklearn.preprocessing import LabelEncoder, OneHotEncoder
labelencoder_X = LabelEncoder()
X[:, 0] = labelencoder_X.fit_transform(X[:, 0])
onehotencoder = OneHotEncoder(categorical_features = [0])
X = onehotencoder.fit_transform(X).toarray()
# Dummy variable trap
X = X[:, 1:]
# Splitting the Dataset into Training Set and Test Set
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2)
# Fitting the Multiple Linear Regression to the training set
""" regressor is an object of LinearRegression() class in line 36 """
from sklearn.linear_model import LinearRegression
regressor = LinearRegression()
regressor.fit(X_train, y_train)
# Predicting the test set result
y_pred = regressor.predict(X_test)