def get_staging_data(root_dir: Path) -> pd.DataFrame:
"""
Given a root registration directory, collate all the staging CSVs into one file.
Write out the combined organ volume CSV into the root registration directory.
Parameters
----------
root_dir
The path to the root registration directory
Returns
-------
The combined dataframe of all the organ volumes
"""
output_dir = root_dir / 'output'
dataframes = []
for line_dir, specimen_dir in specimen_iterator(output_dir):
staging_info = specimen_dir / 'output' / common.STAGING_INFO_FILENAME
if not staging_info.is_file():
raise FileNotFoundError(f'Cannot find staging info file {staging_info}')
df = pd.read_csv(staging_info, index_col=0)
df['line'] = line_dir.name
dataframes.append(df)
# Write the concatenated staging info to the
all_staging = pd.concat(dataframes)
outpath = output_dir / common.STAGING_INFO_FILENAME
all_staging.to_csv(outpath)
return all_staging
def memorymap_data(self, lama_root_dir: Path) -> Dict[str, np.memmap]:
"""
Iterate over output folder getting each ...........
Parameters
----------
lama_root_dir
Returns
-------
"""
imgs = OrderedDict()
for line_dir, spec_dir in specimen_iterator(lama_root_dir):
config_file = common.getfile_endswith(
'.toml') # Get the Lama config from the specimen directory
config = LamaConfig(config_file)
reg_dir = config['root_reg_dir']
basename = os.path.basename(imgpath)
loader = common.LoadImage(imgpath)
if not loader:
logging.error("Problem normalising image: {}".format(
loader.error_msg))
sys.exit()
arr = loader.array
t = tempfile.TemporaryFile()
m = np.memmap(t, dtype=arr.dtype, mode='w+', shape=arr.shape)
m[:] = arr
imgs[basename] = m
return imgs
def move_stuff(root_dir: Path, out_dir: Path, spec_ids: [List]):
for line_dir, spec_dir in specimen_iterator(root_dir):
if spec_dir.name in spec_ids:
print(f"Moving {spec_dir.name}")
dest = out_dir / spec_dir.name
shutil.move(spec_dir, dest)
def get_staging_data(root: Path, line=None) -> pd.DataFrame:
"""
Collate all the staging data from a folder. Include specimens from all lines.
Save a combined csv in the 'output' directory and return as a DataFrame too.
Parameters
----------
root
The root directory to search
line
Only select staging data for this line
"""
output_dir = root / 'output'
dataframes = []
for line_dir, specimen_dir in specimen_iterator(output_dir):
if line and line_dir.name != line:
continue
staging_file = specimen_dir / 'output' / common.STAGING_INFO_FILENAME
if not staging_file.is_file():
raise FileNotFoundError(
f'Cannot find organ volume file {staging_file}')
df = pd.read_csv(staging_file, index_col=0)
df['line'] = line_dir.name
dataframes.append(df)
# Write the concatenated organ vol file to single csv
staging = pd.concat(dataframes)
# Temp fix to deal with old data
# If first column is 1 or 'value', change it to staging
staging.rename(columns={'1': 'staging', 'value': 'staging'}, inplace=True)
outpath = output_dir / common.STAGING_INFO_FILENAME
staging.to_csv(outpath)
return staging
def get_organ_volume_data(root_dir: Path) -> pd.DataFrame:
"""
Given a root registration directory, collate all the organ volume CSVs into one file.
Write out the combined organ volume CSV into the root registration directory.
Parameters
----------
root_dir
The path to the root registration directory
Returns
-------
The combined data frame of all the organ volumes
specimen id in index organs in rows
"""
output_dir = root_dir / 'output'
dataframes = []
for line_dir, specimen_dir in specimen_iterator(output_dir):
organ_vol_file = specimen_dir / 'output' / common.ORGAN_VOLUME_CSV_FILE
if not organ_vol_file.is_file():
raise FileNotFoundError(f'Cannot find organ volume file {organ_vol_file}')
df = pd.read_csv(organ_vol_file, index_col=0)
if len(df) == 0:
raise ValueError(f'{organ_vol_file} is empty')
dataframes.append(df)
# Write the concatenated organ vol file to single csv
all_organs = pd.concat(dataframes)
# outpath = output_dir / common.ORGAN_VOLUME_CSV_FILE
# all_organs.to_csv(outpath)
return all_organs
def _get_organ_volumes(self, root_dir: Path) -> pd.DataFrame:
"""
Given a root registration directory, collate all the organ volume CSVs into one file.
Write out the combined organ volume CSV into the root registration directory.
Parameters
----------
root_dir
The path to the root registration directory
Returns
-------
The combined dataframe of all the organ volumes
"""
output_dir = root_dir / 'output'
dataframes = []
for line_dir, specimen_dir in specimen_iterator(output_dir):
organ_vol_file = specimen_dir / 'output' / common.ORGAN_VOLUME_CSV_FILE
if not organ_vol_file.is_file():
raise FileNotFoundError(
f'Cannot find organ volume file {organ_vol_file}')
df = pd.read_csv(organ_vol_file, index_col=0)
self._drop_empty_columns(df)
df['line'] = line_dir.name
dataframes.append(df)
# Write the concatenated organ vol file to single csv
if not dataframes:
raise ValueError(
f'No data forund in output directory: {output_dir}')
all_organs = pd.concat(dataframes)
return all_organs
import sys
from PyQt5 import QtGui
from vpv.vpv import Vpv
from lama.paths import specimen_iterator
root_dir = Path('/mnt/IMPC_research/neil/E14.5/baselines/output')
app = QtGui.QApplication([])
ex = Vpv()
vols = []
t = 0
for line_dir, spec_dir in specimen_iterator(root_dir):
def_img = spec_dir / 'output' / 'registrations' / 'deformable_50_to_10' / spec_dir.name / (
spec_dir.name + '.nrrd')
if not def_img.is_file():
print(f'{def_img} not found')
continue
if 'xy' in spec_dir.name.lower():
vols.append(def_img)
# if t > 5:
# break
# t +=1
ex.load_volumes(vols, 'vol')
sys.exit(app.exec_())