def get_from_excel(data_path, extra_sheet=None):
'''
This opens a file dialog allowing ot select an excel file containing
the tracked data, and returns a :class:`CellCluster` object.
Paramteters
-----------
data_path: the path to the excelTM file
Returns
-------
cellcluster : a :class:`CellCluster` instance
the container class for the tracking
Notes
-----
The excel file should follow the structure of `excel_trajs_example.xlsx`
in the project's `data` directory
'''
### Read the data
trajs = pd.read_excel(data_path, 0)
trajs.t_stamp = trajs.t_stamp.astype(np.int)
trajs.label = trajs.label.astype(np.int)
trajs.set_index(['t_stamp', 'label'],
inplace=True)
### The Trajectories class is a subclass of
### pandas DataFrame
### Parsing excel files tends to add NaNs to the data
trajs = Trajectories(trajs.dropna().sortlevel())
metadata = pd.read_excel(data_path, 1)
metadata = {name: value for name, value
in zip(metadata['Name'], metadata['Value'])}
metadata['FileName'] = data_path
store_path = metadata['FileName']
if '.' in store_path[-6:]:
store_path = ''.join(store_path.split('.')[:-1]+['.h5'])
else:
store_path = store_path+'.h5'
store_path = os.path.join(
os.path.dirname(data_path), store_path)
### The ObjectsIO class
objectsio = ObjectsIO(metadata=metadata, store_path=store_path)
cellcluster = CellCluster(objectsio=objectsio)
cellcluster.trajs = trajs
cellcluster.oio['trajs'] = trajs
if extra_sheet is not None:
try:
extra = pd.read_excel(data_path, extra_sheet)
cellcluster.extra = extra
cellcluster.oio['extra'] = extra
except:
print('Extra data from sheet {} not found in the file {}'.format(extra_sheet, data_path))
return cellcluster
def load_multiple_excel(data_path, extra_sheet=None):
xlsx_file = pd.io.excel.ExcelFile(data_path)
lastsheet = xlsx_file.book.nsheets - 1
global_metadata = pd.read_excel(data_path, lastsheet)
global_metadata = {name: value for name, value
in zip(global_metadata['Name'],
global_metadata['Value'])}
clusters = {}
global_metadata['FileName'] = global_metadata['FileName'].replace(' ', '')
for i, name in enumerate(global_metadata['FileName'].split(',')):
### Read the data
trajs = pd.read_excel(data_path, i)
trajs.t_stamp = trajs.t_stamp.astype(np.int)
trajs.label = trajs.label.astype(np.int)
trajs.set_index(['t_stamp', 'label'],
inplace=True)
trajs = Trajectories(trajs.dropna())
metadata = global_metadata.copy()
metadata['FileName'] = os.path.join(
os.path.dirname(data_path), name)
store_path = metadata['FileName']
if '.' in store_path[-6:]:
store_path = ''.join(store_path.split('.')[:-1]+['.h5'])
else:
store_path = store_path+'.h5'
store_path = os.path.join(
os.path.dirname(data_path), store_path)
### The ObjectsIO class
objectsio = ObjectsIO(metadata=metadata, store_path=store_path)
cellcluster = CellCluster(objectsio=objectsio)
cellcluster.trajs = trajs
cellcluster.oio['trajs'] = trajs
clusters[name] = cellcluster
return clusters