def SPCA(model_data, components = None, transform_data = None):
t0 = time()
spca = SparsePCA(n_components=components)
if transform_data == None:
projection = spca.fit_transform(model_data)
else:
spca.fit(model_data)
projection = spca.transform(transform_data)
print "Sparse PCA Time: %0.3f" % (time() - t0)
return projection
def SPCA(model_data, components=None, transform_data=None):
t0 = time()
spca = SparsePCA(n_components=components)
if transform_data == None:
projection = spca.fit_transform(model_data)
else:
spca.fit(model_data)
projection = spca.transform(transform_data)
print "Sparse PCA Time: %0.3f" % (time() - t0)
return projection
df = pd.DataFrame(d)
# print df.head()
# df = pd.get_dummies(df,drop_first=True)
X = list(df['features'])
X = np.array(X)
from scipy import sparse
# X=sparse.csr_matrix(X)
# print(b)
from sklearn.decomposition.truncated_svd import TruncatedSVD
from sklearn.decomposition.sparse_pca import SparsePCA
from sklearn.decomposition import dict_learning_online
sparsepca = SparsePCA(n_components=200)
X = sparsepca.fit_transform(X)
pca = TruncatedSVD(n_components=2)
# X = pca.fit_transform(X)
# X = X.reshape(-1, 1)
Y = df['tag']
from sklearn.model_selection import train_test_split
# X_train, X_test, y_train, y_test = train_test_split(X,Y , test_size=0.2, random_state=42,stratify=Y)
X_train, X_test, y_train, y_test = train_test_split(X,
Y,
test_size=0.2,
random_state=42)
#