def _save_dataset(self, X, y):
""" It saves the dataset that will be used to classify
(helpful for plotting).
Parameters
----------
X : pandas dataframe
dataframe of features
length: point features * motion features
y : pandas dataframe
class labels
study : str
name of previous work
can be: "zheng", "dabiri", "xiao", or "jiang"
"""
# Save dataset
create_folder(self.folder_baseline)
point_features = ["mean", "std"] #["min", "max", "mean", "std"]
colnames = list(product(self.motion_features, point_features))
colnames = list(map("_".join, colnames))
colnames = ["classes"] + colnames
data = pd.concat([y, X], axis=1, ignore_index=True)
data.columns = colnames
filename = self.folder_baseline + "DATASET_baseline.csv"
data.to_csv(filename, index=False)
def _save_dataset(self, X, y, parameter):
""" It saves the dataset that will be used to classify
(helpful for plotting).
Parameters
----------
X : pandas dataframe
dataframe of features
y : pandas dataframe
class labels
"""
# Save dataset
create_folder(self.folder_classification)
## get feature name
colnames_op = self._colnames(self.op_features)
colnames = ["classes"] + colnames_op
name = "_op_" + str(parameter) + ".csv"
data = pd.concat([y, X], axis=1, ignore_index=True)
data.columns = colnames
filename = self.folder_classification + "DATASET_" + name
data.to_csv(filename, index=False)
def _save_features(self, df_features, path_to_save, segment):
""" This function saves the features extracted from trajectory.
Parameters
----------
df_features : pandas dataframe
the segment to save
path_to_save : str
absolute path where to save the segments
segment : str
the user's name
Returns
-------
no value
"""
## save data
user_path, file_name = split(segment)
user_name = basename(dirname(user_path))
transportation_name = basename(user_path)
## where to save the features
path_to_save = join(path_to_save, user_name, transportation_name,
file_name)
create_folder(path_to_save)
df_features.to_csv(path_to_save, sep=",", header=True, index=None)
def _save_op_transformation(self, df_op, path_to_save, file_name):
""" Saves the OP transformation.
Parameters
----------
df_op : pandas dataframe
the transformations to save
path_to_save : str
absolute path where to save the df
file_name : str
name to save data
Returns
-------
no value
"""
## save data
path_to_save = join(path_to_save, file_name)
create_folder(path_to_save)
df_op.to_csv(path_to_save, sep=",", header=True, index=None)
def _save_results(self, df, file_name):
""" It saves the classification results (metrics and confusion matrices)
in the self.folder_baseline (it will create the folder, if it doesn't exist).
Parameters
----------
df : pandas dataframe
the results to save
file_name : str
the name to save the result
Returns
-------
no value
"""
create_folder(self.folder_baseline)
path_to_save = join(self.folder_baseline, file_name)
try:
df.to_csv(path_to_save, sep=",")
except: # for confusion matrices
df = np.array(df)
with open(path_to_save, 'w') as outfile:
for enum, data_slice in enumerate(df):
if enum != 10:
# Writing out a break to indicate different slices...
outfile.write(
'# Confusion Matrix for CV{}\n'.format(enum + 1))
else:
outfile.write('# Total Confusion Matrix\n')
np.savetxt(outfile, data_slice, fmt='%d')
def _save_segments(self, transport_name, segment, folder_segments, user_name):
""" Saves the segments extracted from trajectory.
Parameters
----------
transport_name : str
the transportation used in the trajectory
segment : pandas dataframe
the segment to save
path_to_save : str
absolute path where to save the segments
user_name : str
the user's name
Returns
-------
no value
"""
path_to_save = join(folder_segments, user_name)
## write to file
previous_traj = get_files(path_to_save, transport_name, False)
k = len(previous_traj) + 1
## where to save
path_to_save = join(path_to_save, transport_name, "")
create_folder(path_to_save)
## name to save
file_name = "{}_{:03d}.csv".format(transport_name, k)
save_in = join(path_to_save, file_name)
segment.to_csv(save_in, sep = ",", header = True, index = None)