def SVM(X, y, X_ind, y_ind, is_reg=False):
"""Cross Validation and independent set test for Support Vector Machine (SVM)
Arguments:
X (ndarray): Feature data of training and validation set for cross-validation.
m X n matrix, m is the No. of samples, n is the No. of fetures
y (ndarray): Label data of training and validation set for cross-validation.
m-D vector, and m is the No. of samples.
X_ind (ndarray): Feature data of independent test set for independent test.
It has the similar data structure as X.
y_ind (ndarray): Feature data of independent set for for independent test.
It has the similar data structure as y
out (str): The file path for saving the result data.
is_reg (bool, optional): define the model for regression (True) or classification (False) (Default: False)
Returns:
cvs (ndarray): cross-validation results. The shape is (m, ), m is the No. of samples.
inds (ndarray): independent test results. It has similar data structure as cvs.
"""
if is_reg:
folds = KFold(5).split(X)
model = SVR()
else:
folds = StratifiedKFold(5).split(X, y)
model = SVC(probability=True)
cvs = np.zeros(y.shape)
inds = np.zeros(y_ind.shape)
gs = GridSearchCV(model, {
'C': 2.0**np.array([-5, 15]),
'gamma': 2.0**np.array([-15, 5])
},
n_jobs=5)
gs.fit(X, y)
params = gs.best_params_
print(params)
for i, (trained, valided) in enumerate(folds):
model = SVC(probability=True, C=params['C'], gamma=params['gamma'])
model.fit(X[trained], y[trained])
if is_reg:
cvs[valided] = model.predict(X[valided])
inds += model.predict(X_ind)
else:
cvs[valided] = model.predict_proba(X[valided])[:, 1]
inds += model.predict_proba(X_ind)[:, 1]
return cvs, inds / 5
class MySVM:
def __init__(self, name, data, problem_type, load_from_file=False):
self.name = name
self._unpack_data(data)
self.problem_type = problem_type
if load_from_file:
self._load_model(self.name)
else:
if problem_type == 'classification':
self.model = SVC(probability=True)
else:
self.model = SVR()
def find_best_model(self, param_grid, save=False):
search = GridSearchCV(self.model,
param_grid=param_grid,
cv=10,
verbose=1,
n_jobs=-1)
search.fit(self.X_train, self.y_train)
print(search.best_params_)
self.model = search.best_estimator_
if save:
print("saving model")
dump(self.model, f'models/{self.name}.joblib')
return search
def train(self, X, y):
self.model.fit(X, y)
def predict(self, X):
if self.problem_type == "classification":
return self.model.predict_proba(X)
elif self.problem_type == "regression":
return self.model.predict(X).reshape(-1, 1)
def _unpack_data(self, data):
self.X_train = data[0]
self.y_train = data[1]
self.X_test = data[2]
self.y_test = data[3]
def _load_model(self, name):
self.model = load(f'models/{name}.joblib')
ynew = model.predict(Xnew)
print("X=%s, Predicted=%s" % (Xnew[0], ynew[0]))
# In[ ]:
from sklearn.linear_model import LogisticRegression
from sklearn.datasets.samples_generator import make_blobs
# generate 2d classification dataset
X, y = make_blobs(n_samples=100, centers=2, n_features=2, random_state=1)
# fit final model
model = LogisticRegression()
model.fit(X, y)
# new instances where we do not know the answer
Xnew, _ = make_blobs(n_samples=3, centers=2, n_features=2, random_state=1)
# make a prediction
ynew = model.predict_proba(Xnew)
# show the inputs and predicted probabilities
for i in range(len(Xnew)):
print("X=%s, Predicted=%s" % (Xnew[i], ynew[i]))
# In[15]:
#creating model
import pandas
from sklearn import model_selection
from sklearn.linear_model import LogisticRegression
from sklearn.externals import joblib
dataframe = pandas.read_csv(
"/home/seethalprince/cdc/CDC_Intern/Dataset/Data1.csv")
array = dataframe.values
def cv(dataset, summary_df, cddd_model_dir, molbert_model_dir):
df, indices = get_data(dataset)
cddd = InferenceModel(cddd_model_dir) # type: ignore
molbert = MolBertFeaturizer(molbert_model_dir,
embedding_type='average-1-cat-pooled',
max_seq_len=200,
device='cpu') # type: ignore
ecfp = MorganFPFeaturizer(fp_size=2048,
radius=2,
use_counts=True,
use_features=False)
rdkit_norm = PhysChemFeaturizer(normalise=True)
cddd_fn = lambda smiles: cddd.seq_to_emb(smiles)
molbert_fn = lambda smiles: molbert.transform(smiles)[0]
ecfp_fn = lambda smiles: ecfp.transform(smiles)[0]
rdkit_norm_fn = lambda smiles: rdkit_norm.transform(smiles)[0]
for i, (train_idx, valid_idx, test_idx) in enumerate(indices):
train_df = df.iloc[train_idx]
valid_df = df.iloc[valid_idx]
# combine train and valid set as SVMs don't use a validation set, but NNs do.
# this way they use the same amount of data.
train_df = pd.concat([train_df, valid_df])
test_df = df.iloc[test_idx]
fn_combos = [('cddd', cddd_fn), ('molbert', molbert_fn),
('ECFP4', ecfp_fn), ('rdkit_norm', rdkit_norm_fn)]
for feat_name, feat_fn in fn_combos:
train_features = np.vstack([
feat_fn(batch) for batch in batchify(train_df['SMILES'], 256)
])
train_labels = train_df[df.columns[-1]]
test_features = np.vstack(
[feat_fn(batch) for batch in batchify(test_df['SMILES'], 256)])
test_labels = test_df[df.columns[-1]]
mode = summary_df[summary_df['task_name'] ==
dataset].iloc[0]['task_type'].strip()
np.random.seed(i)
if mode == 'regression':
model = SVR(C=5.0)
elif mode == 'classification':
model = SVC(5.0, probability=True)
else:
raise ValueError(
f'Mode has to be either classification or regression but was {mode}.'
)
model.fit(train_features, train_labels)
predictions = model.predict(test_features)
if mode == 'classification':
# predict probabilities (needed for some metrics) and get probs of positive class ([:, 1])
prob_predictions = model.predict_proba(test_features)[:, 1]
metrics_dict = {
'AUROC':
lambda: metrics.roc_auc_score(test_labels, prob_predictions
),
'AveragePrecision':
lambda: metrics.average_precision_score(
test_labels, prob_predictions),
'Accuracy':
lambda: metrics.accuracy_score(test_labels, predictions),
}
else:
metrics_dict = {
'MAE':
lambda: metrics.mean_absolute_error(
test_labels, predictions),
'RMSE':
lambda: np.sqrt(
metrics.mean_squared_error(test_labels, predictions)),
'MSE':
lambda: metrics.mean_squared_error(test_labels, predictions
),
'R2':
lambda: metrics.r2_score(test_labels, predictions),
}
metric_values = {}
for name, callable_metric in metrics_dict.items():
try:
metric_values[name] = callable_metric()
except Exception as e:
print(f'unable to calculate {name} metric')
print(e)
metric_values[name] = np.nan
default_path = os.path.join(
'./logs/',
datetime.now().strftime('%Y-%m-%dT%H:%M:%SZ'))
output_dir = os.path.join(default_path, dataset, str(i))
os.makedirs(output_dir, exist_ok=True)
with open(os.path.join(output_dir, f'{feat_name}_metrics.json'),
'w+') as fp:
json.dump(metric_values, fp)
class Log_reg():
def __init__(self):
self.log_reg = SVR(kernel='rbf',
degree=3,
gamma='auto',
coef0=0.0,
tol=1e-3,
C=1.0,
epsilon=0.1,
shrinking=True,
cache_size=200,
verbose=False,
max_iter=-1)
self.ss = StandardScaler()
def read_features_csv(self, file, My_Model):
# 需要注意第一行数据未读取
df = pd.read_csv(file)
columns_size = df.columns.size
if My_Model == 0:
X = df.iloc[:, 1:columns_size - args.k - 2]
else:
X = pd.concat([
df.iloc[:, 1:columns_size - 2 * args.k - 2],
df.iloc[:, columns_size - args.k - 2:-2]
],
axis=1)
Y = df.iloc[:, -2]
# print(X.head())
# print(X,Y)
return X, Y
# print(df.head())
def fit_logistic_regression(self, train_file):
x, y = self.read_features_csv(train_file, args.myModel)
# ss = StandardScaler()
x = self.ss.fit_transform(x)
# print(x,y)
# log_reg = LogisticRegression()
self.log_reg.fit(x, y)
# y_pre = log_reg.predict_proba(x)
# return log_reg
def predict_test(self, test_file):
x, y = self.read_features_csv(test_file, args.myModel)
x = self.ss.fit_transform(x)
y_p_pre = self.log_reg.predict_proba(x)
y_pre = self.log_reg.predict(x)
accuracy = accuracy_score(y, y_pre)
precision = precision_score(y, y_pre)
recall = recall_score(y, y_pre)
F1 = f1_score(y, y_pre)
scores = self.log_reg.score(x, y)
print("accuracy, precision, recall, F1:", accuracy, precision, recall,
F1)
print("scores:", scores)
return y_p_pre, y
