def main():
X_train, X_valid, y_train, y_valid = colour_predict.data()
depth = 10
leaf = 10
modelrgb = make_pipeline(
MinMaxScaler(),
DecisionTreeClassifier(max_depth=depth, min_samples_leaf=leaf))
modelrgb.fit(X_train, y_train)
print("Model Score RGB Training: ", modelrgb.score(X_train, y_train))
print("Model Score RGB Validation: ", modelrgb.score(X_valid, y_valid))
modellab = make_pipeline(
FunctionTransformer(colour_predict.rgb_to_lab, validate=False),
MinMaxScaler(),
DecisionTreeClassifier(max_depth=depth, min_samples_leaf=leaf))
modellab.fit(X_train, y_train)
print("Model Score LAB Training: ", modellab.score(X_train, y_train))
print("Model Score LAB Validation: ", modellab.score(X_valid, y_valid))
modelhsv = make_pipeline(
FunctionTransformer(colour_predict.rgb_to_hsv, validate=False),
MinMaxScaler(),
DecisionTreeClassifier(max_depth=depth, min_samples_leaf=leaf))
modelhsv.fit(X_train, y_train)
print("Model Score HSV Training: ", modelhsv.score(X_train, y_train))
print("Model Score HSV Validation: ", modelhsv.score(X_valid, y_valid))
modelall = make_pipeline(
FunctionTransformer(colour_predict.get_all_features, validate=False),
MinMaxScaler(),
DecisionTreeClassifier(max_depth=depth, min_samples_leaf=leaf))
modelall.fit(X_train, y_train)
print("Model Score All Features Training: ",
modelall.score(X_train, y_train))
print("Model Score All Features Validation: ",
modelall.score(X_valid, y_valid))
modelallpca = make_pipeline(
FunctionTransformer(colour_predict.get_all_features, validate=False),
MinMaxScaler(), PCA(),
DecisionTreeClassifier(max_depth=depth, min_samples_leaf=leaf))
modelallpca.fit(X_train, y_train)
print("Model Score All Features with PCA Training: ",
modelallpca.score(X_train, y_train))
print("Model Score All Features with PCA Validation: ",
modelallpca.score(X_valid, y_valid))
modelallpoly = make_pipeline(
FunctionTransformer(colour_predict.get_all_features, validate=False),
MinMaxScaler(), PolynomialFeatures(degree=4, include_bias=True),
DecisionTreeClassifier(max_depth=depth, min_samples_leaf=leaf))
modelallpoly.fit(X_train, y_train)
print("Model Score All Features with Polynomial Training: ",
modelallpoly.score(X_train, y_train))
print("Model Score All Features with Polynomial Validation: ",
modelallpoly.score(X_valid, y_valid))
def main():
X_train, X_valid, y_train, y_valid = colour_predict.data()
k = 10
modelrgb = make_pipeline(KNeighborsClassifier(n_neighbors=k))
modelrgb.fit(X_train, y_train)
print("Model Score RGB Training: ", modelrgb.score(X_train, y_train))
print("Model Score RGB Validation: ", modelrgb.score(X_valid, y_valid))
modellab = make_pipeline(
FunctionTransformer(colour_predict.rgb_to_lab, validate=False),
KNeighborsClassifier(n_neighbors=k))
modellab.fit(X_train, y_train)
print("Model Score LAB Training: ", modellab.score(X_train, y_train))
print("Model Score LAB Validation: ", modellab.score(X_valid, y_valid))
modelhsv = make_pipeline(
FunctionTransformer(colour_predict.rgb_to_hsv, validate=False),
KNeighborsClassifier(n_neighbors=k))
modelhsv.fit(X_train, y_train)
print("Model Score HSV Training: ", modelhsv.score(X_train, y_train))
print("Model Score HSV Validation: ", modelhsv.score(X_valid, y_valid))
modelall = make_pipeline(
FunctionTransformer(colour_predict.get_all_features, validate=False),
KNeighborsClassifier(n_neighbors=k))
modelall.fit(X_train, y_train)
print("Model Score All Features Training: ",
modelall.score(X_train, y_train))
print("Model Score All Features Validation: ",
modelall.score(X_valid, y_valid))
modelallpca = make_pipeline(
FunctionTransformer(colour_predict.get_all_features, validate=False),
PCA(), KNeighborsClassifier(n_neighbors=k))
modelallpca.fit(X_train, y_train)
print("Model Score All Features with PCA Training: ",
modelallpca.score(X_train, y_train))
print("Model Score All Features with PCA Validation: ",
modelallpca.score(X_valid, y_valid))
modelallpoly = make_pipeline(
FunctionTransformer(colour_predict.get_all_features, validate=False),
PolynomialFeatures(degree=4, include_bias=True),
KNeighborsClassifier(n_neighbors=k))
modelallpoly.fit(X_train, y_train)
print("Model Score All Features with Polynomial Training: ",
modelallpoly.score(X_train, y_train))
print("Model Score All Features with Polynomial Validation: ",
modelallpoly.score(X_valid, y_valid))
def main():
X_train, X_valid, y_train, y_valid = colour_predict.data()
est = 400
depth = 10
leaf = 10
modelrgb = make_pipeline(
RandomForestClassifier(n_estimators=est, max_depth=depth, min_samples_leaf=leaf)
)
modelrgb.fit(X_train, y_train)
print("Model Score RGB Training: ", modelrgb.score(X_train, y_train))
print("Model Score RGB Validation: ", modelrgb.score(X_valid, y_valid))
modellab = make_pipeline(
FunctionTransformer(colour_predict.rgb_to_lab, validate=False),
RandomForestClassifier(n_estimators=est, max_depth=depth, min_samples_leaf=leaf)
)
modellab.fit(X_train, y_train)
print("Model Score LAB Training: ", modellab.score(X_train, y_train))
print("Model Score LAB Validation: ", modellab.score(X_valid, y_valid))
modelhsv = make_pipeline(
FunctionTransformer(colour_predict.rgb_to_hsv, validate=False),
RandomForestClassifier(n_estimators=est, max_depth=depth, min_samples_leaf=leaf)
)
modelhsv.fit(X_train, y_train)
print("Model Score HSV Training: ", modelhsv.score(X_train, y_train))
print("Model Score HSV Validation: ", modelhsv.score(X_valid, y_valid))
modelall = make_pipeline(
FunctionTransformer(colour_predict.get_all_features, validate=False),
RandomForestClassifier(n_estimators=est, max_depth=depth, min_samples_leaf=leaf)
)
modelall.fit(X_train, y_train)
print("Model Score All Features Training: ", modelall.score(X_train, y_train))
print("Model Score All Features Validation: ", modelall.score(X_valid, y_valid))
modelallpca = make_pipeline(
FunctionTransformer(colour_predict.get_all_features, validate=False),
PCA(),
RandomForestClassifier(n_estimators=est, max_depth=depth, min_samples_leaf=leaf)
)
modelallpca.fit(X_train, y_train)
print("Model Score All Features with PCA Training: ", modelallpca.score(X_train, y_train))
print("Model Score All Features with PCA Validation: ", modelallpca.score(X_valid, y_valid))
def main():
X_train, X_valid, y_train, y_valid = colour_predict.data()
modelrgb = make_pipeline(MinMaxScaler(), GaussianNB())
modelrgb.fit(X_train, y_train)
print("Model Score RGB Training: ", modelrgb.score(X_train, y_train))
print("Model Score RGB Validation: ", modelrgb.score(X_valid, y_valid))
#colour_predict.plot_predictions(modelrgb)
modellab = make_pipeline(
FunctionTransformer(colour_predict.rgb_to_lab, validate=False),
MinMaxScaler(), GaussianNB())
modellab.fit(X_train, y_train)
print("Model Score LAB Training: ", modellab.score(X_train, y_train))
print("Model Score LAB Validation: ", modellab.score(X_valid, y_valid))
#colour_predict.plot_predictions(modellab)
modelhsv = make_pipeline(
FunctionTransformer(colour_predict.rgb_to_hsv, validate=False),
MinMaxScaler(), GaussianNB())
modelhsv.fit(X_train, y_train)
print("Model Score HSV Training: ", modelhsv.score(X_train, y_train))
print("Model Score HSV Validation: ", modelhsv.score(X_valid, y_valid))
#colour_predict.plot_predictions(modelhsv)
modelall = make_pipeline(
FunctionTransformer(colour_predict.get_all_features, validate=False),
MinMaxScaler(), GaussianNB())
modelall.fit(X_train, y_train)
print("Model Score All Features Training: ",
modelall.score(X_train, y_train))
print("Model Score All Features Validation: ",
modelall.score(X_valid, y_valid))
#colour_predict.plot_predictions(modelall)
modelallpca = make_pipeline(
FunctionTransformer(colour_predict.get_all_features, validate=False),
MinMaxScaler(), PCA(), GaussianNB())
modelallpca.fit(X_train, y_train)
print("Model Score All Features with PCA Training: ",
modelallpca.score(X_train, y_train))
print("Model Score All Features with PCA Validation: ",
modelallpca.score(X_valid, y_valid))
def main():
X_train, X_valid, y_train, y_valid = colour_predict.data()
modelrgb = make_pipeline(
MinMaxScaler(),
VotingClassifier([
('nb', GaussianNB()),
('knn', KNeighborsClassifier(3)),
('svm', SVC(kernel='rbf', C=10, gamma=5)),
('tree1', DecisionTreeClassifier(max_depth=10)),
('tree2', DecisionTreeClassifier(max_depth=10, min_samples_leaf=10)),])
)
modelrgb.fit(X_train, y_train)
print("Model Score RGB Training: ", modelrgb.score(X_train, y_train))
print("Model Score RGB Validation: ", modelrgb.score(X_valid, y_valid))
modellab = make_pipeline(
FunctionTransformer(colour_predict.rgb_to_lab, validate=False),
MinMaxScaler(),
VotingClassifier([
('nb', GaussianNB()),
('knn', KNeighborsClassifier(3)),
('svm', SVC(kernel='rbf', C=10, gamma=5)),
('tree1', DecisionTreeClassifier(max_depth=10)),
('tree2', DecisionTreeClassifier(max_depth=10, min_samples_leaf=10)),])
)
modellab.fit(X_train, y_train)
print("Model Score LAB Training: ", modellab.score(X_train, y_train))
print("Model Score LAB Validation: ", modellab.score(X_valid, y_valid))
modelhsv = make_pipeline(
FunctionTransformer(colour_predict.rgb_to_hsv, validate=False),
MinMaxScaler(),
VotingClassifier([
('nb', GaussianNB()),
('knn', KNeighborsClassifier(3)),
('svm', SVC(kernel='rbf', C=10, gamma=5)),
('tree1', DecisionTreeClassifier(max_depth=10)),
('tree2', DecisionTreeClassifier(max_depth=10, min_samples_leaf=10)),])
)
modelhsv.fit(X_train, y_train)
print("Model Score HSV Training: ", modelhsv.score(X_train, y_train))
print("Model Score HSV Validation: ", modelhsv.score(X_valid, y_valid))
modelall = make_pipeline(
FunctionTransformer(colour_predict.get_all_features, validate=False),
MinMaxScaler(),
VotingClassifier([
('nb', GaussianNB()),
('knn', KNeighborsClassifier(3)),
('svm', SVC(kernel='rbf', C=10, gamma=5)),
('tree1', DecisionTreeClassifier(max_depth=10)),
('tree2', DecisionTreeClassifier(max_depth=10, min_samples_leaf=10)),])
)
modelall.fit(X_train, y_train)
print("Model Score All Features Training: ", modelall.score(X_train, y_train))
print("Model Score All Features Validation: ", modelall.score(X_valid, y_valid))
modelallpca = make_pipeline(
FunctionTransformer(colour_predict.get_all_features, validate=False),
MinMaxScaler(),
PCA(),
VotingClassifier([
('nb', GaussianNB()),
('knn', KNeighborsClassifier(3)),
('svm', SVC(kernel='rbf', C=10, gamma=5)),
('tree1', DecisionTreeClassifier(max_depth=10)),
('tree2', DecisionTreeClassifier(max_depth=10, min_samples_leaf=10)),])
)
modelallpca.fit(X_train, y_train)
print("Model Score All Features Training: ", modelallpca.score(X_train, y_train))
print("Model Score All Features Validation: ", modelallpca.score(X_valid, y_valid))