一、Permutation Importance
Permutation importance is calculated after a model has been fitted.
Randomly re-ordering a single column should cause less accurate predictions, since the resulting data no longer corresponds to anything observed in the real world. Model accuracy especially suffers if we shuffle a column that the model relied on heavily for predictions.
the process is as follows:
1.Get a trained model
2.Shuffle the values in a single column, make predictions using the resulting dataset. Use these predictions and the true target values to calculate how much the loss function suffered from shuffling.
3.repeat step 2 with the next column in the dataset, until you have calculated the importance of each column.
Code Example：
loads the data and builds a rudimentary model
import numpy as np
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestClassifier
data = pd.read_csv('../input/fifa-2018-match-statistics/FIFA 2018 Statistics.csv')
y = (data['Man of the Match'] == "Yes") # Convert from string "Yes"/"No" to binary
feature_names = [i for i in data.columns if data[i].dtype in [np.int64]]
X = data[feature_names]
train_X, val_X, train_y, val_y = train_test_split(X, y, random_state=1)
my_model = RandomForestClassifier(random_state=0).fit(train_X, train_y)
to calculate and show importances with the eli5 library:
import eli5
from eli5.sklearn import PermutationImportance
perm = PermutationImportance(my_model, random_state=1).fit(val_X, val_y)
eli5.show_weights(perm, feature_names = val_X.columns.tolist())
Interpreting Permutation Importances
The values towards the top are the most important features, and those towards the bottom matter least.
The first number in each row shows how much model performance decreased with a random shuffling ( on average)
The number after the ± measures how performance varied from one-reshuffling to the next.
二、Partial Plots
feature importance shows what variables most affect predictions.
partial dependence plots show how a feature affects predictions.
partial dependence plots are calculated after a model has been fit.
We will use the fitted model to predict our outcome. But we repeatedly alter the value for one variable to make a series of predictions.
Code Example
model building
import numpy as np
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestClassifier
from sklearn.tree import DecisionTreeClassifier
data = pd.read_csv('../input/fifa-2018-match-statistics/FIFA 2018 Statistics.csv')
y = (data['Man of the Match'] == "Yes") # Convert from string "Yes"/"No" to binary
feature_names = [i for i in data.columns if data[i].dtype in [np.int64]]
X = data[feature_names]
train_X, val_X, train_y, val_y = train_test_split(X, y, random_state=1)
tree_model = DecisionTreeClassifier(random_state=0, max_depth=5, min_samples_split=5).fit(train_X, train_y)
create the Partial Dependence Plot using the PDPBox library.
from matplotlib import pyplot as plt
from pdpbox import pdp, get_dataset, info_plots
# Create the data that we will plot
pdp_goals = pdp.pdp_isolate(model=tree_model, dataset=val_X, model_features=feature_names, feature='Goal Scored')
# plot it
pdp.pdp_plot(pdp_goals, 'Goal Scored')
plt.show()
The y axis is interpreted as change in the prediction from what it would be predicted at the baseline or leftmost value.
A blue shaded area indicates level of confidence
2D Partial Dependence Plots
2D partial dependence plots are useful about interactions between features.
# Similar to previous PDP plot except we use pdp_interact instead of pdp_isolate and pdp_interact_plot instead of pdp_isolate_plot
features_to_plot = ['Goal Scored', 'Distance Covered (Kms)']
inter1 = pdp.pdp_interact(model=tree_model, dataset=val_X, model_features=feature_names, features=features_to_plot)
pdp.pdp_interact_plot(pdp_interact_out=inter1, feature_names=features_to_plot, plot_type='contour')
plt.show()
For example, we see the highest predictions when a team scores at least 1 goal and they run a total distance close to 100km. If they score 0 goals, distance covered doesn’t matter.
三、SHAP Values
SHAP Values break down a prediction to show the impact of each feature.
SHAP values interpret the impact of having a certain value for a given feature in comparison to the prediction we’d make if that feature took some baseline value.
import numpy as np
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestClassifier
data = pd.read_csv('../input/fifa-2018-match-statistics/FIFA 2018 Statistics.csv')
y = (data['Man of the Match'] == "Yes") # Convert from string "Yes"/"No" to binary
feature_names = [i for i in data.columns if data[i].dtype in [np.int64, np.int64]]
X = data[feature_names]
train_X, val_X, train_y, val_y = train_test_split(X, y, random_state=1)
my_model = RandomForestClassifier(random_state=0).fit(train_X, train_y)
look at SHAP values for a single row of the dataset.
row_to_show = 5
data_for_prediction = val_X.iloc[row_to_show] # use 1 row of data here. Could use multiple rows if desired
data_for_prediction_array = data_for_prediction.values.reshape(1, -1)
import shap # package used to calculate Shap values
# Create object that can calculate shap values
explainer = shap.TreeExplainer(my_model)
# Calculate Shap values
shap_values = explainer.shap_values(data_for_prediction)
The shap_values object above is a list with two arrays. The first array is the SHAP values for a negative outcome (don’t win the award), and the second array is the list of SHAP values for the positive outcome (wins the award).
shap.initjs()
shap.force_plot(explainer.expected_value[1], shap_values[1], data_for_prediction)
We predicted 0.7, whereas the base_value is 0.4979.
Feature values causing increased predictions are in pink, and their visual size shows the magnitude of the feature’s effect.
Feature values decreasing the prediction are in blue.
四、Advanced Uses of SHAP Values
Summary Plots
Permutation importance is great because it created simple numeric measures to see which features mattered to a model. But it doesn’t tell you how each features matter.
SHAP summary plots give us a birds-eye view of feature importance and what is driving it.
import numpy as np
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestClassifier
data = pd.read_csv('../input/fifa-2018-match-statistics/FIFA 2018 Statistics.csv')
y = (data['Man of the Match'] == "Yes") # Convert from string "Yes"/"No" to binary
feature_names = [i for i in data.columns if data[i].dtype in [np.int64, np.int64]]
X = data[feature_names]
train_X, val_X, train_y, val_y = train_test_split(X, y, random_state=1)
my_model = RandomForestClassifier(random_state=0).fit(train_X, train_y)
import shap # package used to calculate Shap values
# Create object that can calculate shap values
explainer = shap.TreeExplainer(my_model)
# calculate shap values. This is what we will plot.
# Calculate shap_values for all of val_X rather than a single row, to have more data for plot.
shap_values = explainer.shap_values(val_X)
# Make plot. Index of [1] is explained in text below.
shap.summary_plot(shap_values[1], val_X)
Vertical location shows what feature it is depicting.
Color shows whether that feature was high or low for that row of the dataset.
Horizontal location shows whether the effect of that value caused a higher or lower prediction.
For example, the point in the upper left was for a team that scored few goals, reducing the prediction by 0.25.
The model ignored the Red and Yellow & Red features.
Usually Yellow Card doesn’t affect the prediction, but there is an extreme case where a high value caused a much lower prediction.
High values of Goal scored caused higher predictions, and low values caused low predictions
Calculating SHAP values can be slow.The exception is when using an xgboost model, which SHAP has some optimizations for and which is thus much faster.
SHAP Dependence Contribution Plots
Partial Dependence Plots show how a single feature impacts predictions. But there’s a lot they don’t show.For instance, what is the distribution of effects? Is the effect of having a certain value pretty constant, or does it vary a lot depending on the values of other feaures.
SHAP dependence contribution plots provide a similar insight to PDP’s, but they add a lot more detail.
import shap # package used to calculate Shap values
# Create object that can calculate shap values
explainer = shap.TreeExplainer(my_model)
# calculate shap values. This is what we will plot.
shap_values = explainer.shap_values(X)
# make plot.
shap.dependence_plot('Ball Possession %', shap_values[1], X, interaction_index="Goal Scored")
Each dot represents a row of the data.
The horizontal location is the actual value from the dataset.
the vertical location shows what having that value did to the prediction.
The spread suggests that other features must interact with Ball Possession %.
The color also help shows the interactions.
In general, having the ball increases a team’s chance of having their player win the award. But if they only score one goal, that trend reverses .