X = incomedata.to_numpy()
incomedata_np = incomedata.to_numpy()
X = X[:,0:41]
## Get norminal data
X = np.delete(X,[0,5,16,17,18,29,38],axis=1)
## Missing values
X = SimpleImputer(missing_values='?', strategy='most_frequent').fit_transform(X)
## Convert to continuous int values
##enc = preprocessing.OneHotEncoder()
##X = enc.fit_transform(X)
enc = preprocessing.LabelEncoder()
for i in range(34):
X[:,i] = enc.fit_transform(X[:,i])
#X = X.toarray()
X = X.astype(int)
## Missing values and Convert to int values
Y = incomedata_np[:,[0,5,16,17,18,29,38]]
Y = SimpleImputer(missing_values='?', strategy='most_frequent').fit_transform(Y)
Y = Y.astype(int)
## Get the X, Y for regression
X = np.hstack((X,Y))
## Y = incomedata[:,0]
## save to text
np.savetxt('incomedata/LRtestData',X)
########################################################################################################################
incomedata = pd.read_csv('incomedata/census-income.data',header=None)
X = incomedata.to_numpy()
incomedata_np = incomedata.to_numpy()
mat = df[Ratings_cols].values
mat = SimpleImputer(missing_values=np.nan, strategy='mean').fit_transform(mat)
np.save("UCI_dataset_04", mat)
np.save("UCI_dataset_05",
pd.read_csv("yacht_hydrodynamics.data", header=None, sep="\s+").values)
np.save("UCI_dataset_06",
pd.read_excel("Concrete_Data.xls", header=0).values.astype(float))
np.save("UCI_dataset_07",
pd.read_csv("airfoil_self_noise.dat", header=None, sep="\t").values)
mat = pd.read_csv("communities.data", header=None, sep=",").values[:, 5:]
mat[mat == "?"] = None
mat = mat.astype(float)
mat[mat == None] = np.nan
mat = SimpleImputer(missing_values=np.nan, strategy='mean').fit_transform(mat)
np.save("UCI_dataset_08", mat)
np.save(
"UCI_dataset_09",
pd.read_csv("CASP.csv", header=0)[[
'F1', 'F2', 'F3', 'F4', 'F5', 'F6', 'F7', 'F8', 'F9', 'RMSD'
]].values.astype(float))
np.save(
"UCI_dataset_10",
pd.read_csv("Relation Network (Directed).data", header=None,
sep=",").values[:, 1:].astype(float))
