Machine Learning - Estimators Evaluation
This notebook summarises a simple workflow to compare the performance of different ML estimators for the same dataset (described here).
In [34]:
## ignore warnings
import warnings
warnings.filterwarnings('ignore')
In [35]:
## relevant modules
import pandas as pd
import numpy as np
from sklearn.metrics import classification_report
from yellowbrick.classifier import ClassificationReport
from pathlib import Path
In [36]:
csvdir="C://Users/Seve//workplace/vimfastml/data/"
training_features = pd.read_csv(Path(csvdir, "features_training.csv"), header = None)
training_labels = pd.read_csv(Path(csvdir, "labels_training.csv"), header = None)
test_features = pd.read_csv(Path(csvdir, "features_test.csv"), header = None)
test_labels = pd.read_csv(Path(csvdir, "labels_test.csv"), header = None)
In [37]:
training_features.head(5)
Out[37]:
| 0 | 1 | |
|---|---|---|
| 0 | 1.147382 | 0.026487 |
| 1 | 0.100893 | 0.882746 |
| 2 | -0.041287 | 0.400452 |
| 3 | 0.657740 | -0.385432 |
| 4 | -0.053942 | 0.268567 |
The code below will displayYellowBrick classification report for several estimators.
In [38]:
from sklearn.metrics import f1_score
from sklearn.pipeline import Pipeline
from sklearn.svm import LinearSVC, NuSVC, SVC
from sklearn.neighbors import KNeighborsClassifier
from sklearn.linear_model import LogisticRegressionCV, LogisticRegression, SGDClassifier
from sklearn.ensemble import BaggingClassifier, ExtraTreesClassifier, RandomForestClassifier,GradientBoostingClassifier
from sklearn.metrics import classification_report
models = [
SVC(gamma='auto'), # Support Vector Classification.
SVC(kernel='linear',class_weight="balanced", C=1.0, random_state=0), # Support Vector Classification. The “balanced” mode uses the values of y to automatically adjust weights inversely proportional to class frequencies in the input data as n_samples / (n_classes * np.bincount(y))
KNeighborsClassifier(),
BaggingClassifier(),
GradientBoostingClassifier(random_state=0),
RandomForestClassifier(n_estimators=300)
]
def print_classification_report(X, y, X_test, y_test, estimator):
"""
print scikit-learn classification report
"""
# Instantiate the estimator
model.fit(X, y)
expected = y_test
predicted = model.predict(X_test)
print("{}:\n{}".format(estimator.__class__.__name__,
classification_report(expected, predicted)))
def display_classification_report(X, y, X_test, y_test, estimator):
"""
yb classification report
"""
visualizer = ClassificationReport(estimator,support=True)
visualizer.fit(X, y)
visualizer.score(X_test, y_test)
visualizer.show()
#print_classification_report(training_features, training_labels.values.flatten(),test_features,test_labels.values.flatten(),BaggingClassifier())
for model in models:
display_classification_report(training_features, training_labels.values.flatten(),test_features,test_labels.values.flatten(),model)
In [ ]: