def __init__(self,
random_state=84,
n_estimators=20,
params={'c_k': [2**i for i in range(-7, 7, 2)]},
niterations=10):
self.model = MLTSVM()
self.params = params
self.niterations = niterations
def ML_model_predict(train_x, train_y, test_x, model_name):
print(f"--------train {model_name} model----------")
classifier = None
if model_name == "MLARAM":
classifier = MLARAM(threshold=0.2)
elif model_name == "MLkNN":
classifier = MLkNN()
elif model_name == "BRkNNa":
classifier = BRkNNaClassifier()
elif model_name == "BRkNNb":
classifier = BRkNNbClassifier()
elif model_name == "RF":
classifier = RandomForestClassifier(n_estimators=1000,
random_state=0,
n_jobs=-1)
elif model_name == "MLTSVM":
classifier = MLTSVM(c_k=2**-1)
classifier.fit(train_x, train_y)
prediction = classifier.predict(test_x)
return prediction
ft_OT = pd.read_csv("feature_similarity_chebi_ontology_DiShIn_2.csv",
names=label_names)
ft_OT.rename(mapper=lambda x: x + "_OT", axis=1, inplace=True)
X = np.concatenate((ft_FP, ft_OT), axis=1)
scoring_funcs = {
"hamming loss": hamming_func,
"aiming": aiming_func,
"coverage": coverage_func,
"accuracy": accuracy_func,
"absolute true": absolute_true_func,
} # Keep recorded
parameters = {'c_k': [2**i for i in range(-5, 5)]}
mtsvm = GridSearchCV(MLTSVM(),
param_grid=parameters,
n_jobs=-1,
cv=loocv,
scoring=scoring_funcs,
verbose=3,
refit="absolute true")
mtsvm.fit(X, Y.values)
print(mtsvm.best_score_)
mytuple = (mtsvm, )
to_save = dump(mytuple, filename="mtsvm.joblib")
def classifiers(self):
return [MLTSVM(c_k=2**-4)]
"coverage": coverage_func,
"accuracy": accuracy_func,
"absolute true": absolute_true_func,
} # Keep recorded
parameters = {'k': range(1,11), 's': [0.5, 0.7, 1.0]}
mlknn = GridSearchCV(MLkNN(), param_grid=parameters, n_jobs=-1, cv=loocv,
scoring=scoring_funcs, verbose=3, refit="absolute true")
mlknn.fit(X, Y.values)
print(mlknn.best_score_)
parameters = {'c_k': [2**i for i in range(-5, 5)]}
mtsvm = GridSearchCV(MLTSVM(), param_grid=parameters, n_jobs=-1, cv=loocv,
scoring=scoring_funcs, verbose=3, refit="absolute true")
mtsvm.fit(X, Y.values)
print(mtsvm.best_score_)
parameters = {
'classifier': [LabelPowerset()],
'classifier__classifier': [ExtraTreesClassifier()],
'classifier__classifier__n_estimators': [50, 100, 500, 1000],
'clusterer' : [
NetworkXLabelGraphClusterer(LabelCooccurrenceGraphBuilder(weighted=True, include_self_edges=False), 'louvain'),
NetworkXLabelGraphClusterer(LabelCooccurrenceGraphBuilder(weighted=True, include_self_edges=False), 'lpa')
]
}
""" """""" """""" """""" """""" """""" """""" """""" """""" """""" """""" """"""
mlknn = MLkNN()
vec = TfidfVectorizer(analyzer='word')
pipe = Pipeline(steps=[('vec', vec), ('MLkNN', mlknn)])
parameters = [{
'vec__ngram_range': [(1, 1), (1, 2), (1, 5)],
'vec__max_features': [5000, 10000, 50000, 100000],
'vec__stop_words': ['english', None],
'MLkNN__k': [1, 3, 10, 20, 50],
'MLkNN__s': [0.5, 0.7, 1.0]
}]
nested_cross_val(pipe, parameters, X, y, "MLkNN")
"""""" """""" """""" """""" """""" """""" """""" """""" """""" """""" """""" """
Run MLTSVM
""" """""" """""" """""" """""" """""" """""" """""" """""" """""" """""" """"""
mltsvm = MLTSVM(max_iteration=1000)
vec = TfidfVectorizer(analyzer='word')
pipe = Pipeline(steps=[('vec', vec), ('MLTSVM', mltsvm)])
parameters = [{
'vec__ngram_range': [(1, 1), (1, 2), (1, 5)],
'vec__max_features': [5000, 10000, 50000, 100000],
'vec__stop_words': ['english', None],
'MLTSVM__c_k': [2**i for i in range(-5, 10, 2)]
}]
nested_cross_val(pipe, parameters, X, y, "MLTSVM")
"""""" """""" """""" """""" """""" """""" """""" """""" """""" """""" """""" """
Run MLARAM
""" """""" """""" """""" """""" """""" """""" """""" """""" """""" """""" """"""
mlaram = MLARAM()
vec = TfidfVectorizer(analyzer='word')
pipe = Pipeline(steps=[('vec', vec), ('MLARAM', mlaram)])