def translate(self, file_path, *args, **kwargs):
# Two kinds of files:
# 1. Simple GSF files -> use metadata, data = gsf_read(file_path)
# 2. Native .gwy files -> use the gwyfile package
# I have a notebook that shows how such data can be read.
# Create the .h5 file from the input file
if not isinstance(file_path, (str, unicode)):
raise TypeError('file_path should be a string!')
if not (file_path.endswith('.gsf') or file_path.endswith('.gwy')):
# TODO: Gwyddion is weird, it doesn't append the file extension some times.
# In theory, you could identify the kind of file by looking at the header (line 38 in gsf_read()).
# Ideally the header check should be used instead of the extension check
raise ValueError('file_path must have a .gsf or .gwy extension!')
file_path = path.abspath(file_path)
folder_path, base_name = path.split(file_path)
base_name = base_name[:-4]
h5_path = path.join(folder_path, base_name + '.h5')
if path.exists(h5_path):
remove(h5_path)
self.h5_file = h5py.File(h5_path, 'w')
"""
Setup the global parameters
---------------------------
translator: Gywddion
data_type: depends on file type
GwyddionGSF_<gsf_meta['title']>
or
GwyddionGWY_<gwy_meta['title']>
"""
self.global_parms = generate_dummy_main_parms()
self.global_parms['translator'] = 'Gwyddion'
# Create the measurement group
meas_grp = create_indexed_group(self.h5_file, 'Measurement')
if file_path.endswith('.gsf'):
self._translate_gsf(file_path, meas_grp)
if file_path.endswith('gwy'):
self._translate_gwy(file_path, meas_grp)
write_simple_attrs(self.h5_file, self.global_parms)
return h5_path
def _setupH5(self, image_parms):
"""
Setup the HDF5 file in which to store the data
Due to the structure of the ndata format, we can only create the Measurement and Channel groups here
Parameters
----------
image_parms : dict
Dictionary of parameters
Returns
-------
h5_main : h5py.Dataset
HDF5 Dataset that the images will be written into
h5_mean_spec : h5py.Dataset
HDF5 Dataset that the mean over all positions will be written
into
h5_ronch : h5py.Dataset
HDF5 Dateset that the mean over all Spectroscopic steps will be
written into
"""
root_parms = generate_dummy_main_parms()
root_parms['data_type'] = 'PtychographyData'
# Create the hdf5 data Group
write_simple_attrs(self.h5_f, root_parms)
h5_channels = list()
for meas_parms in image_parms:
# Create new measurement group for each set of parameters
meas_grp = create_indexed_group(self.h5_f, 'Measurement')
# Write the parameters as attributes of the group
write_simple_attrs(meas_grp, meas_parms)
chan_grp = create_indexed_group(meas_grp, 'Channel')
h5_channels.append(chan_grp)
self.h5_f.flush()
return h5_channels
def _setupH5(self, image_parms):
"""
Setup the HDF5 file in which to store the data
Due to the structure of the ndata format, we can only create the Measurement and Channel groups here
Parameters
----------
image_parms : dict
Dictionary of parameters
Returns
-------
h5_main : h5py.Dataset
HDF5 Dataset that the images will be written into
h5_mean_spec : h5py.Dataset
HDF5 Dataset that the mean over all positions will be written
into
h5_ronch : h5py.Dataset
HDF5 Dateset that the mean over all Spectroscopic steps will be
written into
"""
root_parms = generate_dummy_main_parms()
root_parms['data_type'] = 'PtychographyData'
# Create the hdf5 data Group
write_simple_attrs(self.h5_f, root_parms)
h5_channels = list()
for meas_parms in image_parms:
# Create new measurement group for each set of parameters
meas_grp = create_indexed_group(self.h5_f, 'Measurement')
# Write the parameters as attributes of the group
write_simple_attrs(meas_grp, meas_parms)
chan_grp = create_indexed_group(meas_grp, 'Channel')
h5_channels.append(chan_grp)
self.h5_f.flush()
return h5_channels
def _setupH5(self, usize, vsize, data_type, scan_size_x, scan_size_y, image_parms):
"""
Setup the HDF5 file in which to store the data including creating
the Position and Spectroscopic datasets
Parameters
----------
usize : int
Number of pixel columns in the images
vsize : int
Number of pixel rows in the images
data_type : type
Data type to save image as
scan_size_x : int
Number of images in the x dimension
scan_size_y : int
Number of images in the y dimension
image_parms : dict
Dictionary of parameters
Returns
-------
h5_main : h5py.Dataset
HDF5 Dataset that the images will be written into
h5_mean_spec : h5py.Dataset
HDF5 Dataset that the mean over all positions will be written
into
h5_ronch : h5py.Dataset
HDF5 Dateset that the mean over all Spectroscopic steps will be
written into
"""
num_pixels = usize * vsize
num_files = scan_size_x * scan_size_y
root_parms = generate_dummy_main_parms()
root_parms['data_type'] = 'PtychographyData'
main_parms = {'num_images': num_files,
'image_size_u': usize,
'image_size_v': vsize,
'num_pixels': num_pixels,
'translator': 'Ptychography',
'scan_size_x': scan_size_x,
'scan_size_y': scan_size_y}
main_parms.update(image_parms)
# Create the hdf5 data Group
write_simple_attrs(self.h5_f, root_parms)
meas_grp = create_indexed_group(self.h5_f, 'Measurement')
write_simple_attrs(meas_grp, main_parms)
chan_grp = create_indexed_group(meas_grp, 'Channel')
# Build the Position and Spectroscopic Datasets
spec_desc = [Dimension('U', 'pixel', np.arange(usize)),
Dimension('V', 'pixel', np.arange(vsize))]
pos_desc = [Dimension('X', 'pixel', np.arange(scan_size_x)),
Dimension('Y', 'pixel', np.arange(scan_size_y))]
ds_chunking = calc_chunks([num_files, num_pixels],
data_type(0).itemsize,
unit_chunks=(1, num_pixels))
# Allocate space for Main_Data and Pixel averaged Data
h5_main = write_main_dataset(chan_grp, (num_files, num_pixels), 'Raw_Data',
'Intensity', 'a.u.',
pos_desc, spec_desc,
chunks=ds_chunking, dtype=data_type)
h5_ronch= chan_grp.create_dataset('Mean_Ronchigram', shape=[num_pixels], dtype=np.float32)
h5_mean_spec = chan_grp.create_dataset('Spectroscopic_Mean', shape=[num_files], dtype=np.float32)
self.h5_f.flush()
return h5_main, h5_mean_spec, h5_ronch
def translate(self, parm_path):
"""
Basic method that translates .mat data files to a single .h5 file
Parameters
------------
parm_path : string / unicode
Absolute file path of the parameters .mat file.
Returns
----------
h5_path : string / unicode
Absolute path of the translated h5 file
"""
parm_path = path.abspath(parm_path)
(folder_path, file_name) = path.split(parm_path)
(file_name, base_name) = path.split(folder_path)
h5_path = path.join(folder_path, base_name + '.h5')
# Read parameters
print('reading parameter files')
parm_dict, excit_wfm, spec_ind_mat = self.__readparms(parm_path)
parm_dict['data_type'] = 'SPORC'
num_rows = parm_dict['grid_num_rows']
num_cols = parm_dict['grid_num_cols']
num_pix = num_rows * num_cols
# new data format
spec_ind_mat = np.transpose(VALUES_DTYPE(spec_ind_mat))
# Now start creating datasets and populating:
pos_desc = [
Dimension('Y', 'm', np.arange(num_rows)),
Dimension('X', 'm', np.arange(num_cols))
]
ds_pos_ind, ds_pos_val = build_ind_val_dsets(pos_desc,
is_spectral=False)
spec_ind_labels = [
'x index', 'y index', 'loop index', 'repetition index',
'slope index'
]
spec_ind_dict = dict()
for col_ind, col_name in enumerate(spec_ind_labels):
spec_ind_dict[col_name] = (slice(col_ind,
col_ind + 1), slice(None))
ds_spec_inds = VirtualDataset('Spectroscopic_Indices',
INDICES_DTYPE(spec_ind_mat))
ds_spec_inds.attrs['labels'] = spec_ind_dict
ds_spec_vals = VirtualDataset('Spectroscopic_Values', spec_ind_mat)
ds_spec_vals.attrs['labels'] = spec_ind_dict
ds_spec_vals.attrs['units'] = ['V', 'V', '', '', '']
ds_excit_wfm = VirtualDataset('Excitation_Waveform',
np.float32(excit_wfm))
ds_raw_data = VirtualDataset('Raw_Data',
data=[],
maxshape=(num_pix, len(excit_wfm)),
dtype=np.float16,
chunking=(1, len(excit_wfm)),
compression='gzip')
# technically should change the date, etc.
chan_grp = VirtualGroup('Channel_000')
chan_grp.attrs = parm_dict
chan_grp.add_children([
ds_pos_ind, ds_pos_val, ds_spec_inds, ds_spec_vals, ds_excit_wfm,
ds_raw_data
])
global_parms = generate_dummy_main_parms()
global_parms['grid_size_x'] = parm_dict['grid_num_cols']
global_parms['grid_size_y'] = parm_dict['grid_num_rows']
# assuming that the experiment was completed:
global_parms['current_position_x'] = parm_dict['grid_num_cols'] - 1
global_parms['current_position_y'] = parm_dict['grid_num_rows'] - 1
global_parms['data_type'] = parm_dict['data_type']
global_parms['translator'] = 'SPORC'
meas_grp = VirtualGroup('Measurement_000')
meas_grp.add_children([chan_grp])
spm_data = VirtualGroup('')
spm_data.attrs = global_parms
spm_data.add_children([meas_grp])
if path.exists(h5_path):
remove(h5_path)
# Write everything except for the main data.
hdf = HDFwriter(h5_path)
h5_refs = hdf.write(spm_data)
h5_main = get_h5_obj_refs(['Raw_Data'], h5_refs)[0]
# Now doing link_h5_objects_as_attrs:
aux_ds_names = [
'Excitation_Waveform', 'Position_Indices', 'Position_Values',
'Spectroscopic_Indices', 'Spectroscopic_Values'
]
link_h5_objects_as_attrs(h5_main,
get_h5_obj_refs(aux_ds_names, h5_refs))
print('reading raw data now...')
# Now read the raw data files:
pos_ind = 0
for row_ind in range(1, num_rows + 1):
for col_ind in range(1, num_cols + 1):
file_path = path.join(
folder_path,
'result_r' + str(row_ind) + '_c' + str(col_ind) + '.mat')
# print('Working on row {} col {}'.format(row_ind,col_ind))
if path.exists(file_path):
# Load data file
pix_data = loadmat(file_path, squeeze_me=True)
# Take the inverse FFT on 1st dimension
pix_vec = np.fft.ifft(np.fft.ifftshift(pix_data['data']))
# Verified with Matlab - no conjugate required here.
h5_main[pos_ind, :] = np.float16(np.real(pix_vec))
hdf.flush() # flush from memory!
else:
print('File for row {} col {} not found'.format(
row_ind, col_ind))
pos_ind += 1
if (100.0 * pos_ind / num_pix) % 10 == 0:
print('Finished reading {} % of data'.format(
int(100 * pos_ind / num_pix)))
hdf.close()
return h5_path
def _setup_h5(self, data_gen_parms):
"""
Setups up the hdf5 file structure before doing the actual generation
Parameters
----------
data_gen_parms : dict
Dictionary containing the parameters to write to the Measurement Group as attributes
Returns
-------
"""
'''
Build the group structure down to the channel group
'''
# Set up the basic group structure
root_grp = VirtualGroup('')
root_parms = generate_dummy_main_parms()
root_parms['translator'] = 'FAKEBEPS'
root_parms['data_type'] = data_gen_parms['data_type']
root_grp.attrs = root_parms
meas_grp = VirtualGroup('Measurement_')
chan_grp = VirtualGroup('Channel_')
meas_grp.attrs.update(data_gen_parms)
# Create the Position and Spectroscopic datasets for the Raw Data
ds_pos_inds, ds_pos_vals, ds_spec_inds, ds_spec_vals = self._build_ancillary_datasets()
raw_chunking = calc_chunks([self.n_pixels,
self.n_spec_bins],
np.complex64(0).itemsize,
unit_chunks=[1, self.n_bins])
ds_raw_data = VirtualDataset('Raw_Data', data=None,
maxshape=[self.n_pixels, self.n_spec_bins],
dtype=np.complex64,
compression='gzip',
chunking=raw_chunking,
parent=meas_grp)
chan_grp.add_children([ds_pos_inds, ds_pos_vals, ds_spec_inds, ds_spec_vals,
ds_raw_data])
meas_grp.add_children([chan_grp])
root_grp.add_children([meas_grp])
hdf = HDFwriter(self.h5_path)
hdf.delete()
h5_refs = hdf.write(root_grp)
# Delete the MicroDatasets to save memory
del ds_raw_data, ds_spec_inds, ds_spec_vals, ds_pos_inds, ds_pos_vals
# Get the file and Raw_Data objects
h5_raw = get_h5_obj_refs(['Raw_Data'], h5_refs)[0]
h5_chan_grp = h5_raw.parent
# Get the Position and Spectroscopic dataset objects
h5_pos_inds = get_h5_obj_refs(['Position_Indices'], h5_refs)[0]
h5_pos_vals = get_h5_obj_refs(['Position_Values'], h5_refs)[0]
h5_spec_inds = get_h5_obj_refs(['Spectroscopic_Indices'], h5_refs)[0]
h5_spec_vals = get_h5_obj_refs(['Spectroscopic_Values'], h5_refs)[0]
# Link the Position and Spectroscopic datasets as attributes of Raw_Data
link_as_main(h5_raw, h5_pos_inds, h5_pos_vals, h5_spec_inds, h5_spec_vals)
'''
Build the SHO Group
'''
sho_grp = VirtualGroup('Raw_Data-SHO_Fit_', parent=h5_chan_grp.name)
# Build the Spectroscopic datasets for the SHO Guess and Fit
sho_spec_starts = np.where(h5_spec_inds[h5_spec_inds.attrs['Frequency']].squeeze() == 0)[0]
sho_spec_labs = get_attr(h5_spec_inds, 'labels')
ds_sho_spec_inds, ds_sho_spec_vals = build_reduced_spec_dsets(h5_spec_inds,
h5_spec_vals,
keep_dim=sho_spec_labs != 'Frequency',
step_starts=sho_spec_starts)
sho_chunking = calc_chunks([self.n_pixels,
self.n_sho_bins],
sho32.itemsize,
unit_chunks=[1, 1])
ds_sho_fit = VirtualDataset('Fit', data=None,
maxshape=[self.n_pixels, self.n_sho_bins],
dtype=sho32,
compression='gzip',
chunking=sho_chunking,
parent=sho_grp)
ds_sho_guess = VirtualDataset('Guess', data=None,
maxshape=[self.n_pixels, self.n_sho_bins],
dtype=sho32,
compression='gzip',
chunking=sho_chunking,
parent=sho_grp)
sho_grp.add_children([ds_sho_fit, ds_sho_guess, ds_sho_spec_inds, ds_sho_spec_vals])
# Write the SHO group and datasets to the file and delete the MicroDataset objects
h5_sho_refs = hdf.write(sho_grp)
del ds_sho_fit, ds_sho_guess, ds_sho_spec_inds, ds_sho_spec_vals
# Get the dataset handles for the fit and guess
h5_sho_fit = get_h5_obj_refs(['Fit'], h5_sho_refs)[0]
h5_sho_guess = get_h5_obj_refs(['Guess'], h5_sho_refs)[0]
# Get the dataset handles for the SHO Spectroscopic datasets
h5_sho_spec_inds = get_h5_obj_refs(['Spectroscopic_Indices'], h5_sho_refs)[0]
h5_sho_spec_vals = get_h5_obj_refs(['Spectroscopic_Values'], h5_sho_refs)[0]
# Link the Position and Spectroscopic datasets as attributes of the SHO Fit and Guess
link_as_main(h5_sho_fit, h5_pos_inds, h5_pos_vals, h5_sho_spec_inds, h5_sho_spec_vals)
link_as_main(h5_sho_guess, h5_pos_inds, h5_pos_vals, h5_sho_spec_inds, h5_sho_spec_vals)
'''
Build the loop group
'''
loop_grp = VirtualGroup('Fit-Loop_Fit_', parent=h5_sho_fit.parent.name)
# Build the Spectroscopic datasets for the loops
loop_spec_starts = np.where(h5_sho_spec_inds[h5_sho_spec_inds.attrs['DC_Offset']].squeeze() == 0)[0]
loop_spec_labs = get_attr(h5_sho_spec_inds, 'labels')
ds_loop_spec_inds, ds_loop_spec_vals = build_reduced_spec_dsets(h5_sho_spec_inds,
h5_sho_spec_vals,
keep_dim=loop_spec_labs != 'DC_Offset',
step_starts=loop_spec_starts)
# Create the loop fit and guess MicroDatasets
loop_chunking = calc_chunks([self.n_pixels, self.n_loops],
loop_fit32.itemsize,
unit_chunks=[1, 1])
ds_loop_fit = VirtualDataset('Fit', data=None,
maxshape=[self.n_pixels, self.n_loops],
dtype=loop_fit32,
compression='gzip',
chunking=loop_chunking,
parent=loop_grp)
ds_loop_guess = VirtualDataset('Guess', data=None,
maxshape=[self.n_pixels, self.n_loops],
dtype=loop_fit32,
compression='gzip',
chunking=loop_chunking,
parent=loop_grp)
# Add the datasets to the loop group then write it to the file
loop_grp.add_children([ds_loop_fit, ds_loop_guess, ds_loop_spec_inds, ds_loop_spec_vals])
h5_loop_refs = hdf.write(loop_grp)
# Delete the MicroDatasets
del ds_loop_spec_vals, ds_loop_spec_inds, ds_loop_guess, ds_loop_fit
# Get the handles to the datasets
h5_loop_fit = get_h5_obj_refs(['Fit'], h5_loop_refs)[0]
h5_loop_guess = get_h5_obj_refs(['Guess'], h5_loop_refs)[0]
h5_loop_spec_inds = get_h5_obj_refs(['Spectroscopic_Indices'], h5_loop_refs)[0]
h5_loop_spec_vals = get_h5_obj_refs(['Spectroscopic_Values'], h5_loop_refs)[0]
# Link the Position and Spectroscopic datasets to the Loop Guess and Fit
link_as_main(h5_loop_fit, h5_pos_inds, h5_pos_vals, h5_loop_spec_inds, h5_loop_spec_vals)
link_as_main(h5_loop_guess, h5_pos_inds, h5_pos_vals, h5_loop_spec_inds, h5_loop_spec_vals)
self.h5_raw = USIDataset(h5_raw)
self.h5_sho_guess = USIDataset(h5_sho_guess)
self.h5_sho_fit = USIDataset(h5_sho_fit)
self.h5_loop_guess = USIDataset(h5_loop_guess)
self.h5_loop_fit = USIDataset(h5_loop_fit)
self.h5_spec_vals = h5_spec_vals
self.h5_spec_inds = h5_spec_inds
self.h5_sho_spec_inds = h5_sho_spec_inds
self.h5_sho_spec_vals = h5_sho_spec_vals
self.h5_loop_spec_inds = h5_loop_spec_inds
self.h5_loop_spec_vals = h5_loop_spec_vals
self.h5_file = h5_raw.file
return
def _setupH5(self, usize, vsize, data_type, scan_size_x, scan_size_y):
"""
Setup the HDF5 file in which to store the data including creating
the Position and Spectroscopic datasets
Parameters
----------
usize : int
Number of pixel columns in the images
vsize : int
Number of pixel rows in the images
data_type : type
Data type to save image as
scan_size_x : int
Number of images in the x dimension
scan_size_y : int
Number of images in the y dimension
Returns
-------
h5_main : h5py.Dataset
HDF5 Dataset that the images will be written into
h5_mean_spec : h5py.Dataset
HDF5 Dataset that the mean over all positions will be written
into
h5_ronch : h5py.Dataset
HDF5 Dateset that the mean over all Spectroscopic steps will be
written into
"""
num_pixels = usize * vsize
num_files = scan_size_x * scan_size_y
root_parms = generate_dummy_main_parms()
root_parms['data_type'] = 'ImageStackData'
main_parms = {
'num_images': num_files,
'image_size_u': usize,
'image_size_v': vsize,
'num_pixels': num_pixels,
'translator': 'ImageStack',
'scan_size_x': scan_size_x,
'scan_size_y': scan_size_y
}
# Create the hdf5 data Group
write_simple_attrs(self.h5_file, root_parms)
meas_grp = create_indexed_group(self.h5_file, 'Measurement')
write_simple_attrs(meas_grp, main_parms)
chan_grp = create_indexed_group(meas_grp, 'Channel')
# Build the Position and Spectroscopic Datasets
spec_desc = [
Dimension('U', 'pixel', np.arange(usize)),
Dimension('V', 'pixel', np.arange(vsize))
]
pos_desc = [
Dimension('X', 'pixel', np.arange(scan_size_x)),
Dimension('Y', 'pixel', np.arange(scan_size_y))
]
ds_chunking = calc_chunks([num_files, num_pixels],
data_type(0).itemsize,
unit_chunks=(1, num_pixels))
# Allocate space for Main_Data and Pixel averaged Data
h5_main = write_main_dataset(chan_grp, (num_files, num_pixels),
'Raw_Data',
'Intensity',
'a.u.',
pos_desc,
spec_desc,
chunks=ds_chunking,
dtype=data_type)
h5_ronch = meas_grp.create_dataset('Stack_Mean',
data=np.zeros(num_pixels,
dtype=np.float32),
dtype=np.float32)
h5_mean_spec = meas_grp.create_dataset('Image_Means',
data=np.zeros(num_files,
dtype=np.float32),
dtype=np.float32)
self.h5_file.flush()
return h5_main, h5_mean_spec, h5_ronch
def translate(self, file_path, verbose=False, parm_encoding='utf-8'):
"""
Translates the provided file to .h5
Parameters
----------
file_path : String / unicode
Absolute path of the .ibw file
verbose : Boolean (Optional)
Whether or not to show print statements for debugging
parm_encoding : str, optional
Codec to be used to decode the bytestrings into Python strings if needed.
Default 'utf-8'
Returns
-------
h5_path : String / unicode
Absolute path of the .h5 file
"""
file_path = path.abspath(file_path)
# Prepare the .h5 file:
folder_path, base_name = path.split(file_path)
base_name = base_name[:-4]
h5_path = path.join(folder_path, base_name + '.h5')
if path.exists(h5_path):
remove(h5_path)
h5_file = h5py.File(h5_path, 'w')
# Load the ibw file first
ibw_obj = bw.load(file_path)
ibw_wave = ibw_obj.get('wave')
parm_dict = self._read_parms(ibw_wave, parm_encoding)
chan_labels, chan_units = self._get_chan_labels(ibw_wave, parm_encoding)
if verbose:
print('Channels and units found:')
print(chan_labels)
print(chan_units)
# Get the data to figure out if this is an image or a force curve
images = ibw_wave.get('wData')
if images.shape[2] != len(chan_labels):
chan_labels = chan_labels[1:] # for layer 0 null set errors in older AR software
if images.ndim == 3: # Image stack
if verbose:
print('Found image stack of size {}'.format(images.shape))
type_suffix = 'Image'
num_rows = parm_dict['ScanLines']
num_cols = parm_dict['ScanPoints']
images = images.transpose(2, 1, 0) # now ordered as [chan, Y, X] image
images = np.reshape(images, (images.shape[0], -1, 1)) # 3D [chan, Y*X points,1]
pos_desc = [Dimension('X', 'm', np.linspace(0, parm_dict['FastScanSize'], num_cols)),
Dimension('Y', 'm', np.linspace(0, parm_dict['SlowScanSize'], num_rows))]
spec_desc = Dimension('arb', 'a.u.', [1])
else: # single force curve
if verbose:
print('Found force curve of size {}'.format(images.shape))
type_suffix = 'ForceCurve'
images = np.atleast_3d(images) # now [Z, chan, 1]
images = images.transpose((1, 2, 0)) # [chan ,1, Z] force curve
# The data generated above varies linearly. Override.
# For now, we'll shove the Z sensor data into the spectroscopic values.
# Find the channel that corresponds to either Z sensor or Raw:
try:
chan_ind = chan_labels.index('ZSnsr')
spec_data = np.atleast_2d(VALUES_DTYPE(images[chan_ind]))
except ValueError:
try:
chan_ind = chan_labels.index('Raw')
spec_data = np.atleast_2d(VALUES_DTYPE(images[chan_ind]))
except ValueError:
# We don't expect to come here. If we do, spectroscopic values remains as is
spec_data = np.arange(images.shape[2])
pos_desc = Dimension('X', 'm', [1])
spec_desc = Dimension('Z', 'm', spec_data)
# Create measurement group
meas_grp = create_indexed_group(h5_file, 'Measurement')
# Write file and measurement level parameters
global_parms = generate_dummy_main_parms()
global_parms['data_type'] = 'IgorIBW_' + type_suffix
global_parms['translator'] = 'IgorIBW'
write_simple_attrs(h5_file, global_parms)
write_simple_attrs(meas_grp, parm_dict)
# Create Position and spectroscopic datasets
h5_pos_inds, h5_pos_vals = write_ind_val_dsets(meas_grp, pos_desc, is_spectral=False)
h5_spec_inds, h5_spec_vals = write_ind_val_dsets(meas_grp, spec_desc, is_spectral=True)
# Prepare the list of raw_data datasets
for chan_data, chan_name, chan_unit in zip(images, chan_labels, chan_units):
chan_grp = create_indexed_group(meas_grp, 'Channel')
write_main_dataset(chan_grp, np.atleast_2d(chan_data), 'Raw_Data',
chan_name, chan_unit,
None, None,
h5_pos_inds=h5_pos_inds, h5_pos_vals=h5_pos_vals,
h5_spec_inds=h5_spec_inds, h5_spec_vals=h5_spec_vals,
dtype=np.float32)
if verbose:
print('Finished preparing raw datasets')
h5_file.close()
return h5_path
def translate(self, file_path, *args, **kwargs):
"""
Translates a given Bruker / Veeco / Nanoscope AFM derived file to HDF5. Currently handles scans, force curves,
and force-distance maps
Note that this translator was written with a single example file for each modality and may be buggy.
Parameters
----------
file_path : str / unicode
path to data file
Returns
-------
h5_path : str / unicode
path to translated HDF5 file
"""
self.file_path = path.abspath(file_path)
self.meta_data, other_parms = self._extract_metadata()
# These files are weirdly named with extensions such as .001
h5_path = file_path.replace('.', '_') + '.h5'
if path.exists(h5_path):
remove(h5_path)
h5_file = h5py.File(h5_path, 'w')
type_suffixes = ['Image', 'Force_Curve', 'Force_Map']
# 0 - stack of scan images
# 1 - single force curve
# 2 - force map
force_count = 0
image_count = 0
for class_name in self.meta_data.keys():
if 'Ciao force image list' in class_name:
force_count += 1
elif 'Ciao image list' in class_name:
image_count += 1
data_type = 0
if force_count > 0:
if image_count > 0:
data_type = 2
else:
data_type = 1
global_parms = generate_dummy_main_parms()
global_parms['data_type'] = 'Bruker_AFM_' + type_suffixes[data_type]
global_parms['translator'] = 'Bruker_AFM'
write_simple_attrs(h5_file, global_parms)
# too many parameters. Making a dummy group just for the parameters.
h5_parms_grp = h5_file.create_group('Parameters')
# We currently have a dictionary of dictionaries. This needs to be flattened
flat_dict = dict()
for class_name, sub_dict in other_parms.items():
for key, val in sub_dict.items():
flat_dict[class_name + '_' + key] = val
write_simple_attrs(h5_parms_grp, flat_dict)
# Create measurement group
h5_meas_grp = create_indexed_group(h5_file, 'Measurement')
# Call the data specific translation function
trans_funcs = [
self._translate_image_stack, self._translate_force_curve,
self._translate_force_map
]
trans_funcs[data_type](h5_meas_grp)
# wrap up and return path
h5_file.close()
return h5_path
def _setupH5(self, usize, vsize, data_type, num_images, main_parms):
"""
Setup the HDF5 file in which to store the data including creating
the Position and Spectroscopic datasets
Parameters
----------
usize : int
Number of pixel columns in the images
vsize : int
Number of pixel rows in the images
data_type : type
Data type to save image as
num_images : int
Number of images in the movie
main_parms : dict
Returns
-------
h5_main : h5py.Dataset
HDF5 Dataset that the images will be written into
h5_mean_spec : h5py.Dataset
HDF5 Dataset that the mean over all positions will be written
into
h5_ronch : h5py.Dataset
HDF5 Dateset that the mean over all Spectroscopic steps will be
written into
"""
num_pixels = usize * vsize
root_parms = generate_dummy_main_parms()
root_parms['data_type'] = 'PtychographyData'
main_parms['num_images'] = num_images
main_parms['image_size_u'] = usize
main_parms['image_size_v'] = vsize
main_parms['num_pixels'] = num_pixels
main_parms['translator'] = 'Movie'
# Create the hdf5 data Group
write_simple_attrs(self.h5_file, root_parms)
meas_grp = create_indexed_group(self.h5_file, 'Measurement')
write_simple_attrs(meas_grp, main_parms)
chan_grp = create_indexed_group(meas_grp, 'Channel')
# Build the Position and Spectroscopic Datasets
spec_dim = Dimension('Time', 's', np.arange(num_images))
pos_dims = [
Dimension('X', 'a.u.', np.arange(usize)),
Dimension('Y', 'a.u.', np.arange(vsize))
]
ds_chunking = calc_chunks([num_pixels, num_images],
data_type(0).itemsize,
unit_chunks=(num_pixels, 1))
# Allocate space for Main_Data and Pixel averaged Data
h5_main = write_main_dataset(chan_grp, (num_pixels, num_images),
'Raw_Data',
'Intensity',
'a.u.',
pos_dims,
spec_dim,
chunks=ds_chunking,
dtype=data_type)
h5_ronch = meas_grp.create_dataset('Mean_Ronchigram',
data=np.zeros(num_pixels,
dtype=np.float32),
dtype=np.float32)
h5_mean_spec = meas_grp.create_dataset('Spectroscopic_Mean',
data=np.zeros(num_images,
dtype=np.float32),
dtype=np.float32)
self.h5_file.flush()
return h5_main, h5_mean_spec, h5_ronch
def translate(self, parm_path):
"""
The main function that translates the provided file into a .h5 file
Parameters
------------
parm_path : string / unicode
Absolute file path of the parameters .mat file.
Returns
----------
h5_path : string / unicode
Absolute path of the translated h5 file
"""
parm_path = path.abspath(parm_path)
parm_dict, excit_wfm = self._read_parms(parm_path)
self._parse_file_path(parm_path)
num_dat_files = len(self.file_list)
f = open(self.file_list[0], 'rb')
spectrogram_size, count_vals = self._parse_spectrogram_size(f)
print("spectrogram size:", spectrogram_size)
num_pixels = parm_dict['grid_num_rows'] * parm_dict['grid_num_cols']
print('Number of pixels: ', num_pixels)
print('Count Values: ', count_vals)
if (num_pixels + 1) != count_vals:
print(
"Data size does not match number of pixels expected. Cannot continue"
)
# Now start creating datasets and populating:
ds_spec_inds, ds_spec_vals = build_ind_val_dsets(Dimension(
'Bias', 'V', excit_wfm),
is_spectral=True,
verbose=False)
ds_spec_vals.data = np.atleast_2d(
excit_wfm) # The data generated above varies linearly. Override.
pos_desc = [
Dimension('X', 'a.u.', np.arange(parm_dict['grid_num_cols'])),
Dimension('Y', 'a.u.', np.arange(parm_dict['grid_num_rows']))
]
ds_pos_ind, ds_pos_val = build_ind_val_dsets(pos_desc,
is_spectral=False,
verbose=False)
ds_raw_data = VirtualDataset('Raw_Data',
data=[],
maxshape=(ds_pos_ind.shape[0],
spectrogram_size - 5),
dtype=np.complex64,
chunking=(1, spectrogram_size - 5),
compression='gzip')
ds_raw_data.attrs['quantity'] = ['Complex']
aux_ds_names = [
'Position_Indices', 'Position_Values', 'Spectroscopic_Indices',
'Spectroscopic_Values'
]
num_ai_chans = np.int(num_dat_files /
2) # Division by 2 due to real/imaginary
# technically should change the date, etc.
spm_data = VirtualGroup('')
global_parms = generate_dummy_main_parms()
global_parms['data_type'] = 'trKPFM'
global_parms['translator'] = 'trKPFM'
spm_data.attrs = global_parms
meas_grp = VirtualGroup('Measurement_000')
meas_grp.attrs = parm_dict
spm_data.add_children([meas_grp])
hdf = HDFwriter(self.h5_path)
# spm_data.showTree()
hdf.write(spm_data, print_log=False)
self.raw_datasets = list()
for chan_index in range(num_ai_chans):
chan_grp = VirtualGroup(
'{:s}{:03d}'.format('Channel_', chan_index),
'/Measurement_000/')
if chan_index == 0:
chan_grp.attrs = {'Harmonic': 1}
else:
chan_grp.attrs = {'Harmonic': 2}
chan_grp.add_children([
ds_pos_ind, ds_pos_val, ds_spec_inds, ds_spec_vals, ds_raw_data
])
h5_refs = hdf.write(chan_grp, print_log=False)
h5_raw = get_h5_obj_refs(['Raw_Data'], h5_refs)[0]
link_h5_objects_as_attrs(h5_raw,
get_h5_obj_refs(aux_ds_names, h5_refs))
self.raw_datasets.append(h5_raw)
self.raw_datasets.append(h5_raw)
# Now that the N channels have been made, populate them with the actual data....
self._read_data(parm_dict, parm_path, spectrogram_size)
hdf.close()
return self.h5_path
def translate(self, parm_path):
"""
The main function that translates the provided file into a .h5 file
Parameters
------------
parm_path : string / unicode
Absolute file path of the parameters .mat file.
Returns
----------
h5_path : string / unicode
Absolute path of the translated h5 file
"""
parm_path = path.abspath(parm_path)
parm_dict, excit_wfm = self._read_parms(parm_path)
self._parse_file_path(parm_path)
num_dat_files = len(self.file_list)
f = open(self.file_list[0], 'rb')
spectrogram_size, count_vals = self._parse_spectrogram_size(f)
print("Excitation waveform shape: ", excit_wfm.shape)
print("spectrogram size:", spectrogram_size)
num_pixels = parm_dict['grid_num_rows'] * parm_dict['grid_num_cols']
print('Number of pixels: ', num_pixels)
print('Count Values: ', count_vals)
if (num_pixels + 1) != count_vals:
print("Data size does not match number of pixels expected. Cannot continue")
#Find how many channels we have to make
num_ai_chans = num_dat_files // 2 # Division by 2 due to real/imaginary
# Now start creating datasets and populating:
#Start with getting an h5 file
h5_file = h5py.File(self.h5_path)
#First create a measurement group
h5_meas_group = create_indexed_group(h5_file, 'Measurement')
#Set up some parameters that will be written as attributes to this Measurement group
global_parms = generate_dummy_main_parms()
global_parms['data_type'] = 'trKPFM'
global_parms['translator'] = 'trKPFM'
write_simple_attrs(h5_meas_group, global_parms)
write_simple_attrs(h5_meas_group, parm_dict)
#Now start building the position and spectroscopic dimension containers
#There's only one spectroscpoic dimension and two position dimensions
#The excit_wfm only has the DC values without any information on cycles, time, etc.
#What we really need is to add the time component. For every DC step there are some time steps.
num_time_steps = (spectrogram_size-5) //excit_wfm.size
#Let's repeat the excitation so that we get the full vector of same size as the spectrogram
#TODO: Check if this is the norm for this type of dataset
full_spect_val = np.copy(excit_wfm).repeat(num_time_steps)
spec_dims = Dimension('Bias', 'V', full_spect_val)
pos_dims = [Dimension('Cols', 'nm', parm_dict['grid_num_cols']),
Dimension('Rows', 'um', parm_dict['grid_num_rows'])]
self.raw_datasets = list()
for chan_index in range(num_ai_chans):
chan_grp = create_indexed_group(h5_meas_group,'Channel')
if chan_index == 0:
write_simple_attrs(chan_grp,{'Harmonic': 1})
else:
write_simple_attrs(chan_grp,{'Harmonic': 2})
h5_raw = write_main_dataset(chan_grp, # parent HDF5 group
(num_pixels, spectrogram_size - 5),
# shape of Main dataset
'Raw_Data', # Name of main dataset
'Deflection', # Physical quantity contained in Main dataset
'V', # Units for the physical quantity
pos_dims, # Position dimensions
spec_dims, # Spectroscopic dimensions
dtype=np.complex64, # data type / precision
compression='gzip',
chunks=(1, spectrogram_size - 5),
main_dset_attrs={'quantity': 'Complex'})
#h5_refs = hdf.write(chan_grp, print_log=False)
#h5_raw = get_h5_obj_refs(['Raw_Data'], h5_refs)[0]
#link_h5_objects_as_attrs(h5_raw, get_h5_obj_refs(aux_ds_names, h5_refs))
self.raw_datasets.append(h5_raw)
self.raw_datasets.append(h5_raw)
# Now that the N channels have been made, populate them with the actual data....
self._read_data(parm_dict, parm_path, spectrogram_size)
h5_file.file.close()
#hdf.close()
return self.h5_path
def _setupH5(self, usize, vsize, data_type, num_images, main_parms):
"""
Setup the HDF5 file in which to store the data including creating
the Position and Spectroscopic datasets
Parameters
----------
usize : int
Number of pixel columns in the images
vsize : int
Number of pixel rows in the images
data_type : type
Data type to save image as
num_images : int
Number of images in the movie
main_parms : dict
Returns
-------
h5_main : h5py.Dataset
HDF5 Dataset that the images will be written into
h5_mean_spec : h5py.Dataset
HDF5 Dataset that the mean over all positions will be written
into
h5_ronch : h5py.Dataset
HDF5 Dateset that the mean over all Spectroscopic steps will be
written into
"""
num_pixels = usize * vsize
root_parms = generate_dummy_main_parms()
root_parms['data_type'] = 'PtychographyData'
main_parms['num_images'] = num_images
main_parms['image_size_u'] = usize
main_parms['image_size_v'] = vsize
main_parms['num_pixels'] = num_pixels
main_parms['translator'] = 'Movie'
# Create the hdf5 data Group
write_simple_attrs(self.h5_file, root_parms)
meas_grp = create_indexed_group(self.h5_file, 'Measurement')
write_simple_attrs(meas_grp, main_parms)
chan_grp = create_indexed_group(meas_grp, 'Channel')
# Build the Position and Spectroscopic Datasets
spec_dim = Dimension('Time', 's', np.arange(num_images))
pos_dims = [Dimension('X', 'a.u.', np.arange(usize)), Dimension('Y', 'a.u.', np.arange(vsize))]
ds_chunking = calc_chunks([num_pixels, num_images],
data_type(0).itemsize,
unit_chunks=(num_pixels, 1))
# Allocate space for Main_Data and Pixel averaged Data
h5_main = write_main_dataset(chan_grp, (num_pixels, num_images), 'Raw_Data',
'Intensity', 'a.u.',
pos_dims, spec_dim,
chunks=ds_chunking, dtype=data_type)
h5_ronch = meas_grp.create_dataset('Mean_Ronchigram',
data=np.zeros(num_pixels, dtype=np.float32),
dtype=np.float32)
h5_mean_spec = meas_grp.create_dataset('Spectroscopic_Mean',
data=np.zeros(num_images, dtype=np.float32),
dtype=np.float32)
self.h5_file.flush()
return h5_main, h5_mean_spec, h5_ronch
def translate(self, file_path, show_plots=True, save_plots=True, do_histogram=False):
"""
Basic method that translates .dat data file(s) to a single .h5 file
Inputs:
file_path -- Absolute file path for one of the data files.
It is assumed that this file is of the OLD data format.
Outputs:
Nothing
"""
file_path = path.abspath(file_path)
(folder_path, basename) = path.split(file_path)
(basename, path_dict) = self._parse_file_path(file_path)
h5_path = path.join(folder_path, basename + '.h5')
if path.exists(h5_path):
remove(h5_path)
self.h5_file = h5py.File(h5_path, 'w')
isBEPS = True
parm_dict = self.__getParmsFromOldMat(path_dict['old_mat_parms'])
ignored_plt_grps = ['in-field'] # Here we assume that there is no in-field.
# If in-field data is captured then the translator would have to be modified.
# Technically, we could do away with this if statement, as isBEPS is always true for this translation
if isBEPS:
parm_dict['data_type'] = 'BEPSData'
std_expt = parm_dict['VS_mode'] != 'load user defined VS Wave from file'
if not std_expt:
warn('This translator does not handle user defined voltage spectroscopy')
return
spec_label = getSpectroscopicParmLabel(parm_dict['VS_mode'])
# Check file sizes:
if 'read_real' in path_dict.keys():
real_size = path.getsize(path_dict['read_real'])
imag_size = path.getsize(path_dict['read_imag'])
else:
real_size = path.getsize(path_dict['write_real'])
imag_size = path.getsize(path_dict['write_imag'])
if real_size != imag_size:
raise ValueError("Real and imaginary file sizes DON'T match!. Ending")
num_rows = int(parm_dict['grid_num_rows'])
num_cols = int(parm_dict['grid_num_cols'])
num_pix = num_rows * num_cols
tot_bins = real_size / (num_pix * 4) # Finding bins by simple division of entire datasize
# Check for case where only a single pixel is missing.
check_bins = real_size / ((num_pix - 1) * 4)
if tot_bins % 1 and check_bins % 1:
warn('Aborting! Some parameter appears to have changed in-between')
return
elif not tot_bins % 1:
# Everything's ok
pass
elif not check_bins % 1:
tot_bins = check_bins
warn('Warning: A pixel seems to be missing from the data. File will be padded with zeros.')
tot_bins = int(tot_bins)
(bin_inds, bin_freqs, bin_FFT, ex_wfm, dc_amp_vec) = self.__readOldMatBEvecs(path_dict['old_mat_parms'])
"""
Because this is the old data format and there is a discrepancy in the number of bins (they seem to be 2 less
than the actual number), we need to re-calculate it based on the available data. This is done below.
"""
band_width = parm_dict['BE_band_width_[Hz]'] * (0.5 - parm_dict['BE_band_edge_trim'])
st_f = parm_dict['BE_center_frequency_[Hz]'] - band_width
en_f = parm_dict['BE_center_frequency_[Hz]'] + band_width
bin_freqs = np.linspace(st_f, en_f, len(bin_inds), dtype=np.float32)
# Forcing standardized datatypes:
bin_inds = np.int32(bin_inds)
bin_freqs = np.float32(bin_freqs)
bin_FFT = np.complex64(bin_FFT)
ex_wfm = np.float32(ex_wfm)
self.FFT_BE_wave = bin_FFT
(UDVS_labs, UDVS_units, UDVS_mat) = self.__buildUDVSTable(parm_dict)
# Remove the unused plot group columns before proceeding:
(UDVS_mat, UDVS_labs, UDVS_units) = trimUDVS(UDVS_mat, UDVS_labs, UDVS_units, ignored_plt_grps)
spec_inds = np.zeros(shape=(2, tot_bins), dtype=INDICES_DTYPE)
# Will assume that all excitation waveforms have same number of bins
# Here, the denominator is 2 because only out of field measruements. For IF + OF, should be 1
num_actual_udvs_steps = UDVS_mat.shape[0] / 2
bins_per_step = tot_bins / num_actual_udvs_steps
# Some more checks
if bins_per_step % 1:
warn('Non integer number of bins per step!')
return
else:
bins_per_step = int(bins_per_step)
num_actual_udvs_steps = int(num_actual_udvs_steps)
stind = 0
for step_index in range(UDVS_mat.shape[0]):
if UDVS_mat[step_index, 2] < 1E-3: # invalid AC amplitude
continue # skip
spec_inds[0, stind:stind + bins_per_step] = np.arange(bins_per_step, dtype=INDICES_DTYPE) # Bin step
spec_inds[1, stind:stind + bins_per_step] = step_index * np.ones(bins_per_step,
dtype=INDICES_DTYPE) # UDVS step
stind += bins_per_step
del stind, step_index
# Some very basic information that can help the processing / analysis crew
parm_dict['num_bins'] = tot_bins
parm_dict['num_pix'] = num_pix
parm_dict['num_udvs_steps'] = num_actual_udvs_steps
global_parms = generate_dummy_main_parms()
global_parms['grid_size_x'] = parm_dict['grid_num_cols']
global_parms['grid_size_y'] = parm_dict['grid_num_rows']
global_parms['experiment_date'] = parm_dict['File_date_and_time']
# assuming that the experiment was completed:
global_parms['current_position_x'] = parm_dict['grid_num_cols'] - 1
global_parms['current_position_y'] = parm_dict['grid_num_rows'] - 1
global_parms['data_type'] = parm_dict['data_type'] # self.__class__.__name__
global_parms['translator'] = 'ODF'
write_simple_attrs(self.h5_file, global_parms)
# Create Measurement and Channel groups
meas_grp = create_indexed_group(self.h5_file, 'Measurement')
write_simple_attrs(meas_grp, parm_dict)
chan_grp = create_indexed_group(meas_grp, 'Channel')
chan_grp.attrs['Channel_Input'] = parm_dict['IO_Analog_Input_1']
# Create Auxilliary Datasets
h5_ex_wfm = chan_grp.create_dataset('Excitation_Waveform', data=ex_wfm)
udvs_slices = dict()
for col_ind, col_name in enumerate(UDVS_labs):
udvs_slices[col_name] = (slice(None), slice(col_ind, col_ind + 1))
h5_UDVS = chan_grp.create_dataset('UDVS',
data=UDVS_mat,
dtype=np.float32)
write_simple_attrs(h5_UDVS, {'labels': UDVS_labs, 'units': UDVS_units})
h5_bin_steps = chan_grp.create_dataset('Bin_Steps',
data=np.arange(bins_per_step, dtype=np.uint32),
dtype=np.uint32)
# Need to add the Bin Waveform type - infer from UDVS
exec_bin_vec = self.signal_type * np.ones(len(bin_inds), dtype=np.int32)
h5_wfm_typ = chan_grp.create_dataset('Bin_Wfm_Type',
data=exec_bin_vec,
dtype=np.int32)
h5_bin_inds = chan_grp.create_dataset('Bin_Indices',
data=bin_inds,
dtype=np.uint32)
h5_bin_freq = chan_grp.create_dataset('Bin_Frequencies',
data=bin_freqs,
dtype=np.float32)
h5_bin_FFT = chan_grp.create_dataset('Bin_FFT',
data=bin_FFT,
dtype=np.complex64)
# Noise floor should be of shape: (udvs_steps x 3 x positions)
h5_noise_floor = chan_grp.create_dataset('Noise_Floor',
shape=(num_pix, num_actual_udvs_steps),
dtype=nf32,
chunks=(1, num_actual_udvs_steps))
"""
ONLY ALLOCATING SPACE FOR MAIN DATA HERE!
Chunk by each UDVS step - this makes it easy / quick to:
1. read data for a single UDVS step from all pixels
2. read an entire / multiple pixels at a time
The only problem is that a typical UDVS step containing 50 steps occupies only 400 bytes.
This is smaller than the recommended chunk sizes of 10,000 - 999,999 bytes
meaning that the metadata would be very substantial.
This assumption is fine since we almost do not handle any user defined cases
"""
"""
New Method for chunking the Main_Data dataset. Chunking is now done in N-by-N squares of UDVS steps by pixels.
N is determined dinamically based on the dimensions of the dataset. Currently it is set such that individual
chunks are less than 10kB in size.
Chris Smith -- [email protected]
"""
pos_dims = [Dimension('X', 'nm', num_cols), Dimension('Y', 'nm', num_rows)]
# Create Spectroscopic Values and Spectroscopic Values Labels datasets
spec_vals, spec_inds, spec_vals_labs, spec_vals_units, spec_vals_names = createSpecVals(UDVS_mat, spec_inds,
bin_freqs,
exec_bin_vec,
parm_dict, UDVS_labs,
UDVS_units)
spec_dims = list()
for row_ind, row_name in enumerate(spec_vals_labs):
spec_dims.append(Dimension(row_name,
spec_vals_units[row_ind],
spec_vals[row_ind]))
pixel_chunking = maxReadPixels(10240, num_pix * num_actual_udvs_steps,
bins_per_step, np.dtype('complex64').itemsize)
chunking = np.floor(np.sqrt(pixel_chunking))
chunking = max(1, chunking)
chunking = min(num_actual_udvs_steps, num_pix, chunking)
self.h5_main = write_main_dataset(chan_grp, (num_pix, tot_bins), 'Raw_Data',
'Piezoresponse', 'V',
pos_dims, spec_dims,
dtype=np.complex64,
chunks=(chunking, chunking * bins_per_step),
compression='gzip')
self.mean_resp = np.zeros(shape=(self.ds_main.shape[1]), dtype=np.complex64)
self.max_resp = np.zeros(shape=(self.ds_main.shape[0]), dtype=np.float32)
self.min_resp = np.zeros(shape=(self.ds_main.shape[0]), dtype=np.float32)
# Now read the raw data files:
self._read_data(path_dict['read_real'], path_dict['read_imag'], parm_dict)
self.h5_file.flush()
generatePlotGroups(self.ds_main, self.mean_resp, folder_path, basename, self.max_resp,
self.min_resp, max_mem_mb=self.max_ram, spec_label=spec_label, show_plots=show_plots,
save_plots=save_plots, do_histogram=do_histogram)
self.h5_file.close()
return h5_path
def translate(self, file_path):
"""
The main function that translates the provided file into a .h5 file
Parameters
----------
file_path : String / unicode
Absolute path of any file in the directory
Returns
-------
h5_path : String / unicode
Absolute path of the h5 file
"""
file_path = path.abspath(file_path)
# Figure out the basename of the data:
(basename, parm_paths, data_paths) = super(GTuneTranslator, self)._parse_file_path(file_path)
(folder_path, unused) = path.split(file_path)
h5_path = path.join(folder_path, basename + '.h5')
if path.exists(h5_path):
remove(h5_path)
# Load parameters from .mat file
matread = loadmat(parm_paths['parm_mat'],
variable_names=['AI_wave', 'BE_wave_AO_0', 'BE_wave_AO_1', 'BE_wave_train',
'BE_wave', 'total_cols', 'total_rows'])
be_wave = np.float32(np.squeeze(matread['BE_wave']))
be_wave_train = np.float32(np.squeeze(matread['BE_wave_train']))
num_cols = int(matread['total_cols'][0][0])
expected_rows = int(matread['total_rows'][0][0])
self.points_per_pixel = len(be_wave)
self.points_per_line = len(be_wave_train)
# Load parameters from .txt file - 'BE_center_frequency_[Hz]', 'IO rate'
is_beps, parm_dict = parmsToDict(parm_paths['parm_txt'])
# Get file byte size:
# For now, assume that bigtime_00 always exists and is the main file
file_size = path.getsize(data_paths[0])
# Calculate actual number of lines since the first few lines may not be saved
self.num_rows = 1.0 * file_size / (4 * self.points_per_pixel * num_cols)
if self.num_rows % 1:
warn('Error - File has incomplete rows')
return None
else:
self.num_rows = int(self.num_rows)
samp_rate = parm_dict['IO_rate_[Hz]']
ex_freq_nominal = parm_dict['BE_center_frequency_[Hz]']
# method 1 for calculating the correct excitation frequency:
pixel_duration = 1.0 * self.points_per_pixel / samp_rate
num_periods = pixel_duration * ex_freq_nominal
ex_freq_correct = 1 / (pixel_duration / np.floor(num_periods))
# correcting the excitation frequency - will be VERY useful during analysis and filtering
parm_dict['BE_center_frequency_[Hz]'] = ex_freq_correct
# Some very basic information that can help the processing crew
parm_dict['points_per_line'] = self.points_per_line
parm_dict['num_bins'] = self.points_per_pixel
parm_dict['grid_num_rows'] = self.num_rows
parm_dict['data_type'] = 'G_mode_line'
if self.num_rows != expected_rows:
print('Note: {} of {} lines found in data file'.format(self.num_rows, expected_rows))
# Calculate number of points to read per line:
self.__bytes_per_row__ = int(file_size / self.num_rows)
# First finish writing all global parameters, create the file too:
h5_file = h5py.File(h5_path, 'w')
global_parms = generate_dummy_main_parms()
global_parms['data_type'] = 'G_mode_line'
global_parms['translator'] = 'G_mode_line'
write_simple_attrs(h5_file, global_parms)
# Next create the Measurement and Channel groups and write the appropriate parameters to them
meas_grp = create_indexed_group(h5_file, 'Measurement')
write_simple_attrs(meas_grp, parm_dict)
# Now that the file has been created, go over each raw data file:
"""
We only allocate the space for the main data here.
This does NOT change with each file. The data written to it does.
The auxiliary datasets will not change with each raw data file since
only one excitation waveform is used
"""
pos_desc = Dimension('Y', 'm', np.arange(self.num_rows))
spec_desc = Dimension('Excitation', 'V', np.tile(VALUES_DTYPE(be_wave), num_cols))
h5_pos_ind, h5_pos_val = write_ind_val_dsets(meas_grp, pos_desc, is_spectral=False)
h5_spec_inds, h5_spec_vals = write_ind_val_dsets(meas_grp, spec_desc, is_spectral=True)
for f_index in data_paths.keys():
chan_grp = create_indexed_group(meas_grp, 'Channel')
h5_main = write_main_dataset(chan_grp, (self.num_rows, self.points_per_pixel * num_cols), 'Raw_Data',
'Deflection', 'V',
None, None,
h5_pos_inds=h5_pos_ind, h5_pos_vals=h5_pos_val,
h5_spec_inds=h5_spec_inds, h5_spec_vals=h5_spec_vals,
chunks=(1, self.points_per_pixel), dtype=np.float16)
# Now transfer scan data in the dat file to the h5 file:
super(GTuneTranslator, self)._read_data(data_paths[f_index], h5_main)
h5_file.close()
print('G-Tune translation complete!')
return h5_path
def translate(self, parm_path):
"""
Basic method that translates .mat data files to a single .h5 file
Parameters
------------
parm_path : string / unicode
Absolute file path of the parameters .mat file.
Returns
----------
h5_path : string / unicode
Absolute path of the translated h5 file
"""
self.parm_path = path.abspath(parm_path)
(folder_path, file_name) = path.split(parm_path)
(file_name, base_name) = path.split(folder_path)
h5_path = path.join(folder_path, base_name + '.h5')
# Read parameters
parm_dict = readGmodeParms(parm_path)
# Add the w^2 specific parameters to this list
parm_data = loadmat(parm_path, squeeze_me=True, struct_as_record=True)
#freq_sweep_parms = parm_data['freqSweepParms']
#parm_dict['freq_sweep_delay'] = np.float(freq_sweep_parms['delay'].item())
gen_sig = parm_data['genSig']
#parm_dict['wfm_fix_d_fast'] = np.int32(gen_sig['restrictT'].item())
#freq_array = np.float32(parm_data['freqArray'])
# prepare and write spectroscopic values
samp_rate = parm_dict['IO_down_samp_rate_[Hz]']
num_bins = int(parm_dict['wfm_n_cycles'] * parm_dict['wfm_p_slow'] *
samp_rate)
w_vec = np.arange(-0.5 * samp_rate, 0.5 * samp_rate,
np.float32(samp_rate / num_bins))
# There is most likely a more elegant solution to this but I don't have the time... Maybe np.meshgrid
spec_val_mat = np.zeros((len(freq_array) * num_bins, 2),
dtype=VALUES_DTYPE)
spec_val_mat[:, 0] = np.tile(w_vec, len(freq_array))
spec_val_mat[:, 1] = np.repeat(freq_array, num_bins)
spec_ind_mat = np.zeros((2, len(freq_array) * num_bins),
dtype=np.int32)
spec_ind_mat[0, :] = np.tile(np.arange(num_bins), len(freq_array))
spec_ind_mat[1, :] = np.repeat(np.arange(len(freq_array)), num_bins)
num_rows = parm_dict['grid_num_rows']
num_cols = parm_dict['grid_num_cols']
parm_dict['data_type'] = 'GVS'
num_pix = num_rows * num_cols
global_parms = generate_dummy_main_parms()
global_parms['grid_size_x'] = parm_dict['grid_num_cols']
global_parms['grid_size_y'] = parm_dict['grid_num_rows']
# assuming that the experiment was completed:
global_parms['current_position_x'] = parm_dict['grid_num_cols'] - 1
global_parms['current_position_y'] = parm_dict['grid_num_rows'] - 1
global_parms['data_type'] = parm_dict[
'data_type'] # self.__class__.__name__
global_parms['translator'] = 'GVS'
# Now start creating datasets and populating:
if path.exists(h5_path):
remove(h5_path)
h5_f = h5py.File(h5_path, 'w')
write_simple_attrs(h5_f, global_parms)
meas_grp = create_indexed_group(h5_f, 'Measurement')
chan_grp = create_indexed_group(meas_grp, 'Channel')
write_simple_attrs(chan_grp, parm_dict)
pos_dims = [
Dimension('X', 'nm', num_rows),
Dimension('Y', 'nm', num_cols)
]
spec_dims = [
Dimension('Response Bin', 'a.u.', num_bins),
Dimension('Excitation Frequency ', 'Hz', len(freq_array))
]
# Minimize file size to the extent possible.
# DAQs are rated at 16 bit so float16 should be most appropriate.
# For some reason, compression is more effective on time series data
h5_main = write_main_dataset(chan_grp, (num_pix, num_bins),
'Raw_Data',
'Deflection',
'V',
pos_dims,
spec_dims,
chunks=(1, num_bins),
dtype=np.float32)
h5_ex_freqs = chan_grp.create_dataset('Excitation_Frequencies',
freq_array)
h5_bin_freq = chan_grp.create_dataset('Bin_Frequencies', w_vec)
# Now doing link_h5_objects_as_attrs:
link_h5_objects_as_attrs(h5_main, [h5_ex_freqs, h5_bin_freq])
# Now read the raw data files:
pos_ind = 0
for row_ind in range(1, num_rows + 1):
for col_ind in range(1, num_cols + 1):
file_path = path.join(
folder_path,
'fSweep_r' + str(row_ind) + '_c' + str(col_ind) + '.mat')
print('Working on row {} col {}'.format(row_ind, col_ind))
if path.exists(file_path):
# Load data file
pix_data = loadmat(file_path, squeeze_me=True)
pix_mat = pix_data['AI_mat']
# Take the inverse FFT on 2nd dimension
pix_mat = np.fft.ifft(np.fft.ifftshift(pix_mat, axes=1),
axis=1)
# Verified with Matlab - no conjugate required here.
pix_vec = pix_mat.transpose().reshape(pix_mat.size)
h5_main[pos_ind, :] = np.float32(pix_vec)
h5_f.flush() # flush from memory!
else:
print('File not found for: row {} col {}'.format(
row_ind, col_ind))
pos_ind += 1
if (100.0 * pos_ind / num_pix) % 10 == 0:
print('completed translating {} %'.format(
int(100 * pos_ind / num_pix)))
h5_f.close()
return h5_path
def translate(self, file_path):
"""
The main function that translates the provided file into a .h5 file
Parameters
----------
file_path : String / unicode
Absolute path of any file in the directory
Returns
-------
h5_path : String / unicode
Absolute path of the h5 file
"""
file_path = path.abspath(file_path)
# Figure out the basename of the data:
(basename, parm_paths, data_paths) = self._parse_file_path(file_path)
(folder_path, unused) = path.split(file_path)
h5_path = path.join(folder_path, basename+'.h5')
if path.exists(h5_path):
remove(h5_path)
# Load parameters from .mat file - 'BE_wave', 'FFT_BE_wave', 'total_cols', 'total_rows'
matread = loadmat(parm_paths['parm_mat'], variable_names=['BE_wave', 'FFT_BE_wave', 'total_cols', 'total_rows'])
be_wave = np.float32(np.squeeze(matread['BE_wave']))
# Need to take the complex conjugate if reading from a .mat file
# FFT_BE_wave = np.conjugate(np.complex64(np.squeeze(matread['FFT_BE_wave'])))
num_cols = int(matread['total_cols'][0][0])
expected_rows = int(matread['total_rows'][0][0])
self.points_per_pixel = len(be_wave)
# Load parameters from .txt file - 'BE_center_frequency_[Hz]', 'IO rate'
is_beps, parm_dict = parmsToDict(parm_paths['parm_txt'])
# Get file byte size:
# For now, assume that bigtime_00 always exists and is the main file
file_size = path.getsize(data_paths[0])
# Calculate actual number of lines since the first few lines may not be saved
self.num_rows = 1.0 * file_size / (4 * self.points_per_pixel * num_cols)
if self.num_rows % 1:
warn('Error - File has incomplete rows')
return None
else:
self.num_rows = int(self.num_rows)
samp_rate = parm_dict['IO_rate_[Hz]']
ex_freq_nominal = parm_dict['BE_center_frequency_[Hz]']
# method 1 for calculating the correct excitation frequency:
pixel_duration = 1.0 * self.points_per_pixel / samp_rate
num_periods = pixel_duration * ex_freq_nominal
ex_freq_correct = 1 / (pixel_duration / np.floor(num_periods))
# method 2 for calculating the exact excitation frequency:
"""
fft_ex_wfm = np.abs(np.fft.fftshift(np.fft.fft(be_wave)))
w_vec = np.linspace(-0.5 * samp_rate, 0.5 * samp_rate - 1.0*samp_rate / self.points_per_pixel,
self.points_per_pixel)
hot_bins = np.squeeze(np.argwhere(fft_ex_wfm > 1E+3))
ex_freq_correct = w_vec[hot_bins[-1]]
"""
# correcting the excitation frequency - will be VERY useful during analysis and filtering
parm_dict['BE_center_frequency_[Hz]'] = ex_freq_correct
# Some very basic information that can help the processing crew
parm_dict['num_bins'] = self.points_per_pixel
parm_dict['grid_num_rows'] = self.num_rows
parm_dict['data_type'] = 'G_mode_line'
if self.num_rows != expected_rows:
print('Note: {} of {} lines found in data file'.format(self.num_rows, expected_rows))
# Calculate number of points to read per line:
self.__bytes_per_row__ = int(file_size/self.num_rows)
# First finish writing all global parameters, create the file too:
h5_f = h5py.File(h5_path, 'w')
global_parms = generate_dummy_main_parms()
global_parms['data_type'] = 'G_mode_line'
global_parms['translator'] = 'G_mode_line'
write_simple_attrs(h5_f, global_parms)
meas_grp = create_indexed_group(h5_f, 'Measurement')
write_simple_attrs(meas_grp, parm_dict)
pos_desc = Dimension('Y', 'm', np.arange(self.num_rows))
spec_desc = Dimension('Excitation', 'V', np.tile(VALUES_DTYPE(be_wave), num_cols))
first_dat = True
for key in data_paths.keys():
# Now that the file has been created, go over each raw data file:
# 1. write all ancillary data. Link data. 2. Write main data sequentially
""" We only allocate the space for the main data here.
This does NOT change with each file. The data written to it does.
The auxiliary datasets will not change with each raw data file since
only one excitation waveform is used"""
chan_grp = create_indexed_group(meas_grp, 'Channel')
if first_dat:
if len(data_paths) > 1:
# All positions and spectra are shared between channels
h5_pos_inds, h5_pos_vals = write_ind_val_dsets(meas_grp, pos_desc, is_spectral=False)
h5_spec_inds, h5_spec_vals = write_ind_val_dsets(meas_grp, spec_desc, is_spectral=True)
elif len(data_paths) == 1:
h5_pos_inds, h5_pos_vals = write_ind_val_dsets(chan_grp, pos_desc, is_spectral=False)
h5_spec_inds, h5_spec_vals = write_ind_val_dsets(chan_grp, spec_desc, is_spectral=True)
first_dat = False
else:
pass
h5_main = write_main_dataset(chan_grp, (self.num_rows, self.points_per_pixel * num_cols), 'Raw_Data',
'Deflection', 'V',
None, None,
h5_pos_inds=h5_pos_inds, h5_pos_vals=h5_pos_vals,
h5_spec_inds=h5_spec_inds, h5_spec_vals=h5_spec_vals,
chunks=(1, self.points_per_pixel), dtype=np.float16)
# Now transfer scan data in the dat file to the h5 file:
self._read_data(data_paths[key], h5_main)
h5_f.close()
print('G-Line translation complete!')
return h5_path
def translate(self, data_filepath, show_plots=True, save_plots=True, do_histogram=False, debug=False):
"""
The main function that translates the provided file into a .h5 file
Parameters
----------------
data_filepath : String / unicode
Absolute path of the data file (.dat)
show_plots : Boolean (Optional. Default is True)
Whether or not to show plots
save_plots : Boolean (Optional. Default is True)
Whether or not to save the generated plots
do_histogram : Boolean (Optional. Default is False)
Whether or not to generate and save 2D histograms of the raw data
debug : Boolean (Optional. default is false)
Whether or not to print log statements
Returns
--------------
h5_path : String / unicode
Absolute path of the generated .h5 file
"""
data_filepath = path.abspath(data_filepath)
# Read the parameter files
self.debug = debug
if debug:
print('BEndfTranslator: Getting file paths')
parm_filepath, udvs_filepath, parms_mat_path = self._parse_file_path(data_filepath)
if debug:
print('BEndfTranslator: Reading Parms text file')
isBEPS, self.parm_dict = parmsToDict(parm_filepath)
self.parm_dict['data_type'] = 'BEPSData'
if not isBEPS:
warn('This is NOT a BEPS new-data-format dataset!')
return None
""" Find out if this is a custom experiment and whether in and out of field were acquired
For a standard experiment where only in / out field is acquired, zeros are stored
even for those UDVS steps without band excitation"""
self.field_mode = self.parm_dict['VS_measure_in_field_loops']
expt_type = self.parm_dict['VS_mode']
self.spec_label = getSpectroscopicParmLabel(expt_type)
std_expt = expt_type in ['DC modulation mode', 'current mode']
self.halve_udvs_steps = False
ignored_plt_grps = []
if std_expt and self.field_mode != 'in and out-of-field':
self.halve_udvs_steps = True
if self.field_mode == 'out-of-field':
ignored_plt_grps = ['in-field']
else:
ignored_plt_grps = ['out-of-field']
h5_path = path.join(self.folder_path, self.basename + '.h5')
if path.exists(h5_path):
remove(h5_path)
if debug:
print('BEndfTranslator: Preparing to read parms.mat file')
self.BE_wave, self.BE_wave_rev, self.BE_bin_inds = self.__get_excit_wfm(parms_mat_path)
if debug:
print('BEndfTranslator: About to read UDVS file')
self.udvs_labs, self.udvs_units, self.udvs_mat = self.__read_udvs_table(udvs_filepath)
# Remove the unused plot group columns before proceeding:
self.udvs_mat, self.udvs_labs, self.udvs_units = trimUDVS(self.udvs_mat, self.udvs_labs, self.udvs_units,
ignored_plt_grps)
if debug:
print('BEndfTranslator: Read UDVS file')
self.num_udvs_steps = self.udvs_mat.shape[0]
# This is absolutely crucial for reconstructing the data chronologically
self.excit_type_vec = (self.udvs_mat[:, 4]).astype(int)
# First figure out how many waveforms are present in the data from the UDVS
unique_waves = self.__get_unique_wave_types(self.excit_type_vec)
self.__unique_waves__ = unique_waves
self.__num_wave_types__ = len(unique_waves)
# print self.__num_wave_types__, 'different excitation waveforms in this experiment'
if debug:
print('BEndfTranslator: Preparing to set up parsers')
# Preparing objects to parse the file(s)
parsers = self.__assemble_parsers()
# Gathering some basic details before parsing the files:
self.max_pixels = parsers[0].get_num_pixels()
s_pixels = np.array(parsers[0].get_spatial_pixels())
self.pos_labels = ['Laser Spot', 'Z', 'Y', 'X']
self.pos_labels = [self.pos_labels[i] for i in np.where(s_pixels > 1)[0]]
self.pos_mat = make_indices_matrix(s_pixels[np.argwhere(s_pixels > 1)].squeeze())
self.pos_units = ['um' for _ in range(len(self.pos_labels))]
# self.pos_mat = np.int32(self.pos_mat)
# Helping Eric out a bit. Remove this section at a later time:
main_parms = generate_dummy_main_parms()
# main_parms['grid_size_x'] = self.parm_dict['grid_num_cols']
# main_parms['grid_size_y'] = self.parm_dict['grid_num_rows']
main_parms['grid_size_x'] = self.parm_dict['grid_num_rows']
main_parms['grid_size_y'] = self.parm_dict['grid_num_cols']
main_parms['experiment_date'] = self.parm_dict['File_date_and_time']
# assuming that the experiment was completed:
main_parms['current_position_x'] = self.parm_dict['grid_num_rows'] - 1
main_parms['current_position_y'] = self.parm_dict['grid_num_cols'] - 1
main_parms['data_type'] = 'BEPSData'
main_parms['translator'] = 'NDF'
# Writing only the root now:
spm_data = VirtualGroup('')
spm_data.attrs = main_parms
self.hdf = HDFwriter(h5_path)
# self.hdf.clear()
# cacheSettings = self.hdf.file.id.get_access_plist().get_cache()
self.hdf.write(spm_data)
########################################################
# Reading and parsing the .dat file(s)
self._read_data(parsers, unique_waves, show_plots, save_plots, do_histogram)
self.hdf.close()
return h5_path
def translate(self, parm_path):
"""
The main function that translates the provided file into a .h5 file
Parameters
------------
parm_path : string / unicode
Absolute file path of the parameters .mat file.
Returns
----------
h5_path : string / unicode
Absolute path of the translated h5 file
"""
parm_path = path.abspath(parm_path)
parm_dict, excit_wfm = self._read_parms(parm_path)
folder_path, base_name = path.split(parm_path)
waste, base_name = path.split(folder_path)
# Until a better method is provided....
with h5py.File(path.join(folder_path, 'line_1.mat'), 'r') as h5_mat_line_1:
num_ai_chans = h5_mat_line_1['data'].shape[1]
h5_path = path.join(folder_path, base_name+'.h5')
if path.exists(h5_path):
remove(h5_path)
with h5py.File(h5_path) as h5_f:
h5_meas_grp = create_indexed_group(h5_f, 'Measurement')
global_parms = generate_dummy_main_parms()
global_parms.update({'data_type': 'gIV', 'translator': 'gIV'})
write_simple_attrs(h5_meas_grp, global_parms)
# Only prepare the instructions for the dimensions here
spec_dims = Dimension('Bias', 'V', excit_wfm)
pos_dims = Dimension('Y', 'm', np.linspace(0, parm_dict['grid_scan_height_[m]'],
parm_dict['grid_num_rows']))
self.raw_datasets = list()
for chan_index in range(num_ai_chans):
h5_chan_grp = create_indexed_group(h5_meas_grp, 'Channel')
write_simple_attrs(h5_chan_grp, parm_dict)
"""
Minimize file size to the extent possible.
DAQs are rated at 16 bit so float16 should be most appropriate.
For some reason, compression is effective only on time series data
"""
h5_raw = write_main_dataset(h5_chan_grp, (parm_dict['grid_num_rows'], excit_wfm.size), 'Raw_Data',
'Current',
'1E-{} A'.format(parm_dict['IO_amplifier_gain']), pos_dims, spec_dims,
dtype=np.float16, chunks=(1, excit_wfm.size), compression='gzip')
self.raw_datasets.append(h5_raw)
# Now that the N channels have been made, populate them with the actual data....
self._read_data(parm_dict, folder_path)
return h5_path
def _setup_h5(self, data_gen_parms):
"""
Setups up the hdf5 file structure before doing the actual generation
Parameters
----------
data_gen_parms : dict
Dictionary containing the parameters to write to the Measurement Group as attributes
Returns
-------
"""
'''
Build the group structure down to the channel group
'''
# Set up the basic group structure
root_grp = VirtualGroup('')
root_parms = generate_dummy_main_parms()
root_parms['translator'] = 'FAKEBEPS'
root_parms['data_type'] = data_gen_parms['data_type']
root_grp.attrs = root_parms
meas_grp = VirtualGroup('Measurement_')
chan_grp = VirtualGroup('Channel_')
meas_grp.attrs.update(data_gen_parms)
# Create the Position and Spectroscopic datasets for the Raw Data
ds_pos_inds, ds_pos_vals, ds_spec_inds, ds_spec_vals = self._build_ancillary_datasets(
)
raw_chunking = calc_chunks([self.n_pixels, self.n_spec_bins],
np.complex64(0).itemsize,
unit_chunks=[1, self.n_bins])
ds_raw_data = VirtualDataset(
'Raw_Data',
data=None,
maxshape=[self.n_pixels, self.n_spec_bins],
dtype=np.complex64,
compression='gzip',
chunking=raw_chunking,
parent=meas_grp)
chan_grp.add_children([
ds_pos_inds, ds_pos_vals, ds_spec_inds, ds_spec_vals, ds_raw_data
])
meas_grp.add_children([chan_grp])
root_grp.add_children([meas_grp])
hdf = HDFwriter(self.h5_path)
hdf.delete()
h5_refs = hdf.write(root_grp)
# Delete the MicroDatasets to save memory
del ds_raw_data, ds_spec_inds, ds_spec_vals, ds_pos_inds, ds_pos_vals
# Get the file and Raw_Data objects
h5_raw = get_h5_obj_refs(['Raw_Data'], h5_refs)[0]
h5_chan_grp = h5_raw.parent
# Get the Position and Spectroscopic dataset objects
h5_pos_inds = get_h5_obj_refs(['Position_Indices'], h5_refs)[0]
h5_pos_vals = get_h5_obj_refs(['Position_Values'], h5_refs)[0]
h5_spec_inds = get_h5_obj_refs(['Spectroscopic_Indices'], h5_refs)[0]
h5_spec_vals = get_h5_obj_refs(['Spectroscopic_Values'], h5_refs)[0]
# Link the Position and Spectroscopic datasets as attributes of Raw_Data
link_as_main(h5_raw, h5_pos_inds, h5_pos_vals, h5_spec_inds,
h5_spec_vals)
'''
Build the SHO Group
'''
sho_grp = VirtualGroup('Raw_Data-SHO_Fit_', parent=h5_chan_grp.name)
# Build the Spectroscopic datasets for the SHO Guess and Fit
sho_spec_starts = np.where(
h5_spec_inds[h5_spec_inds.attrs['Frequency']].squeeze() == 0)[0]
sho_spec_labs = get_attr(h5_spec_inds, 'labels')
ds_sho_spec_inds, ds_sho_spec_vals = build_reduced_spec_dsets(
h5_spec_inds,
h5_spec_vals,
keep_dim=sho_spec_labs != 'Frequency',
step_starts=sho_spec_starts)
sho_chunking = calc_chunks([self.n_pixels, self.n_sho_bins],
sho32.itemsize,
unit_chunks=[1, 1])
ds_sho_fit = VirtualDataset('Fit',
data=None,
maxshape=[self.n_pixels, self.n_sho_bins],
dtype=sho32,
compression='gzip',
chunking=sho_chunking,
parent=sho_grp)
ds_sho_guess = VirtualDataset(
'Guess',
data=None,
maxshape=[self.n_pixels, self.n_sho_bins],
dtype=sho32,
compression='gzip',
chunking=sho_chunking,
parent=sho_grp)
sho_grp.add_children(
[ds_sho_fit, ds_sho_guess, ds_sho_spec_inds, ds_sho_spec_vals])
# Write the SHO group and datasets to the file and delete the MicroDataset objects
h5_sho_refs = hdf.write(sho_grp)
del ds_sho_fit, ds_sho_guess, ds_sho_spec_inds, ds_sho_spec_vals
# Get the dataset handles for the fit and guess
h5_sho_fit = get_h5_obj_refs(['Fit'], h5_sho_refs)[0]
h5_sho_guess = get_h5_obj_refs(['Guess'], h5_sho_refs)[0]
# Get the dataset handles for the SHO Spectroscopic datasets
h5_sho_spec_inds = get_h5_obj_refs(['Spectroscopic_Indices'],
h5_sho_refs)[0]
h5_sho_spec_vals = get_h5_obj_refs(['Spectroscopic_Values'],
h5_sho_refs)[0]
# Link the Position and Spectroscopic datasets as attributes of the SHO Fit and Guess
link_as_main(h5_sho_fit, h5_pos_inds, h5_pos_vals, h5_sho_spec_inds,
h5_sho_spec_vals)
link_as_main(h5_sho_guess, h5_pos_inds, h5_pos_vals, h5_sho_spec_inds,
h5_sho_spec_vals)
'''
Build the loop group
'''
loop_grp = VirtualGroup('Fit-Loop_Fit_', parent=h5_sho_fit.parent.name)
# Build the Spectroscopic datasets for the loops
loop_spec_starts = np.where(h5_sho_spec_inds[
h5_sho_spec_inds.attrs['DC_Offset']].squeeze() == 0)[0]
loop_spec_labs = get_attr(h5_sho_spec_inds, 'labels')
ds_loop_spec_inds, ds_loop_spec_vals = build_reduced_spec_dsets(
h5_sho_spec_inds,
h5_sho_spec_vals,
keep_dim=loop_spec_labs != 'DC_Offset',
step_starts=loop_spec_starts)
# Create the loop fit and guess MicroDatasets
loop_chunking = calc_chunks([self.n_pixels, self.n_loops],
loop_fit32.itemsize,
unit_chunks=[1, 1])
ds_loop_fit = VirtualDataset('Fit',
data=None,
maxshape=[self.n_pixels, self.n_loops],
dtype=loop_fit32,
compression='gzip',
chunking=loop_chunking,
parent=loop_grp)
ds_loop_guess = VirtualDataset('Guess',
data=None,
maxshape=[self.n_pixels, self.n_loops],
dtype=loop_fit32,
compression='gzip',
chunking=loop_chunking,
parent=loop_grp)
# Add the datasets to the loop group then write it to the file
loop_grp.add_children(
[ds_loop_fit, ds_loop_guess, ds_loop_spec_inds, ds_loop_spec_vals])
h5_loop_refs = hdf.write(loop_grp)
# Delete the MicroDatasets
del ds_loop_spec_vals, ds_loop_spec_inds, ds_loop_guess, ds_loop_fit
# Get the handles to the datasets
h5_loop_fit = get_h5_obj_refs(['Fit'], h5_loop_refs)[0]
h5_loop_guess = get_h5_obj_refs(['Guess'], h5_loop_refs)[0]
h5_loop_spec_inds = get_h5_obj_refs(['Spectroscopic_Indices'],
h5_loop_refs)[0]
h5_loop_spec_vals = get_h5_obj_refs(['Spectroscopic_Values'],
h5_loop_refs)[0]
# Link the Position and Spectroscopic datasets to the Loop Guess and Fit
link_as_main(h5_loop_fit, h5_pos_inds, h5_pos_vals, h5_loop_spec_inds,
h5_loop_spec_vals)
link_as_main(h5_loop_guess, h5_pos_inds, h5_pos_vals,
h5_loop_spec_inds, h5_loop_spec_vals)
self.h5_raw = USIDataset(h5_raw)
self.h5_sho_guess = USIDataset(h5_sho_guess)
self.h5_sho_fit = USIDataset(h5_sho_fit)
self.h5_loop_guess = USIDataset(h5_loop_guess)
self.h5_loop_fit = USIDataset(h5_loop_fit)
self.h5_spec_vals = h5_spec_vals
self.h5_spec_inds = h5_spec_inds
self.h5_sho_spec_inds = h5_sho_spec_inds
self.h5_sho_spec_vals = h5_sho_spec_vals
self.h5_loop_spec_inds = h5_loop_spec_inds
self.h5_loop_spec_vals = h5_loop_spec_vals
self.h5_file = h5_raw.file
return
def translate(self, parm_path):
"""
Basic method that translates .mat data files to a single .h5 file
Parameters
------------
parm_path : string / unicode
Absolute file path of the parameters .mat file.
Returns
----------
h5_path : string / unicode
Absolute path of the translated h5 file
"""
parm_path = path.abspath(parm_path)
(folder_path, file_name) = path.split(parm_path)
(file_name, base_name) = path.split(folder_path)
h5_path = path.join(folder_path, base_name + '.h5')
# Read parameters
print('reading parameter files')
parm_dict, excit_wfm, spec_ind_mat = self.__readparms(parm_path)
parm_dict['data_type'] = 'SPORC'
num_rows = parm_dict['grid_num_rows']
num_cols = parm_dict['grid_num_cols']
num_pix = num_rows * num_cols
# new data format
spec_ind_mat = np.transpose(VALUES_DTYPE(spec_ind_mat))
# Now start creating datasets and populating:
pos_desc = [Dimension('Y', 'm', np.arange(num_rows)), Dimension('X', 'm', np.arange(num_cols))]
ds_pos_ind, ds_pos_val = build_ind_val_dsets(pos_desc, is_spectral=False)
spec_ind_labels = ['x index', 'y index', 'loop index', 'repetition index', 'slope index']
spec_ind_dict = dict()
for col_ind, col_name in enumerate(spec_ind_labels):
spec_ind_dict[col_name] = (slice(col_ind, col_ind + 1), slice(None))
ds_spec_inds = VirtualDataset('Spectroscopic_Indices', INDICES_DTYPE(spec_ind_mat))
ds_spec_inds.attrs['labels'] = spec_ind_dict
ds_spec_vals = VirtualDataset('Spectroscopic_Values', spec_ind_mat)
ds_spec_vals.attrs['labels'] = spec_ind_dict
ds_spec_vals.attrs['units'] = ['V', 'V', '', '', '']
ds_excit_wfm = VirtualDataset('Excitation_Waveform', np.float32(excit_wfm))
ds_raw_data = VirtualDataset('Raw_Data', data=[],
maxshape=(num_pix, len(excit_wfm)),
dtype=np.float16, chunking=(1, len(excit_wfm)),
compression='gzip')
# technically should change the date, etc.
chan_grp = VirtualGroup('Channel_000')
chan_grp.attrs = parm_dict
chan_grp.add_children([ds_pos_ind, ds_pos_val, ds_spec_inds, ds_spec_vals,
ds_excit_wfm, ds_raw_data])
global_parms = generate_dummy_main_parms()
global_parms['grid_size_x'] = parm_dict['grid_num_cols']
global_parms['grid_size_y'] = parm_dict['grid_num_rows']
# assuming that the experiment was completed:
global_parms['current_position_x'] = parm_dict['grid_num_cols'] - 1
global_parms['current_position_y'] = parm_dict['grid_num_rows'] - 1
global_parms['data_type'] = parm_dict['data_type']
global_parms['translator'] = 'SPORC'
meas_grp = VirtualGroup('Measurement_000')
meas_grp.add_children([chan_grp])
spm_data = VirtualGroup('')
spm_data.attrs = global_parms
spm_data.add_children([meas_grp])
if path.exists(h5_path):
remove(h5_path)
# Write everything except for the main data.
hdf = HDFwriter(h5_path)
h5_refs = hdf.write(spm_data)
h5_main = get_h5_obj_refs(['Raw_Data'], h5_refs)[0]
# Now doing link_h5_objects_as_attrs:
aux_ds_names = ['Excitation_Waveform', 'Position_Indices', 'Position_Values',
'Spectroscopic_Indices', 'Spectroscopic_Values']
link_h5_objects_as_attrs(h5_main, get_h5_obj_refs(aux_ds_names, h5_refs))
print('reading raw data now...')
# Now read the raw data files:
pos_ind = 0
for row_ind in range(1, num_rows + 1):
for col_ind in range(1, num_cols + 1):
file_path = path.join(folder_path, 'result_r' + str(row_ind) + '_c' + str(col_ind) + '.mat')
# print('Working on row {} col {}'.format(row_ind,col_ind))
if path.exists(file_path):
# Load data file
pix_data = loadmat(file_path, squeeze_me=True)
# Take the inverse FFT on 1st dimension
pix_vec = np.fft.ifft(np.fft.ifftshift(pix_data['data']))
# Verified with Matlab - no conjugate required here.
h5_main[pos_ind, :] = np.float16(np.real(pix_vec))
hdf.flush() # flush from memory!
else:
print('File for row {} col {} not found'.format(row_ind, col_ind))
pos_ind += 1
if (100.0 * pos_ind / num_pix) % 10 == 0:
print('Finished reading {} % of data'.format(int(100 * pos_ind / num_pix)))
hdf.close()
return h5_path
def translate(self, file_path, verbose=False, append_path='',
grp_name='Measurement', parm_encoding='utf-8'):
"""
Translates the provided file to .h5
Parameters
----------
file_path : String / unicode
Absolute path of the .ibw file
verbose : Boolean (Optional)
Whether or not to show print statements for debugging
append_path : string (Optional)
h5_file to add these data to, must be a path to the h5_file on disk
grp_name : string (Optional)
Change from default "Measurement" name to something specific
parm_encoding : str, optional
Codec to be used to decode the bytestrings into Python strings if needed.
Default 'utf-8'
Returns
-------
h5_path : String / unicode
Absolute path of the .h5 file
"""
file_path = path.abspath(file_path)
# Prepare the .h5 file:
folder_path, base_name = path.split(file_path)
base_name = base_name[:-4]
if not append_path:
h5_path = path.join(folder_path, base_name + '.h5')
if path.exists(h5_path):
remove(h5_path)
h5_file = h5py.File(h5_path, 'w')
else:
h5_path = append_path
if not path.exists(append_path):
raise Exception('File does not exist. Check pathname.')
h5_file = h5py.File(h5_path, 'r+')
# Load the ibw file first
ibw_obj = bw.load(file_path)
ibw_wave = ibw_obj.get('wave')
parm_dict = self._read_parms(ibw_wave, parm_encoding)
chan_labels, chan_units = self._get_chan_labels(ibw_wave, parm_encoding)
if verbose:
print('Channels and units found:')
print(chan_labels)
print(chan_units)
# Get the data to figure out if this is an image or a force curve
images = ibw_wave.get('wData')
if images.shape[-1] != len(chan_labels):
chan_labels = chan_labels[1:] # for layer 0 null set errors in older AR software
if images.ndim == 3: # Image stack
if verbose:
print('Found image stack of size {}'.format(images.shape))
type_suffix = 'Image'
num_rows = parm_dict['ScanLines']
num_cols = parm_dict['ScanPoints']
images = images.transpose(2, 1, 0) # now ordered as [chan, Y, X] image
images = np.reshape(images, (images.shape[0], -1, 1)) # 3D [chan, Y*X points,1]
pos_desc = [Dimension('X', 'm', np.linspace(0, parm_dict['FastScanSize'], num_cols)),
Dimension('Y', 'm', np.linspace(0, parm_dict['SlowScanSize'], num_rows))]
spec_desc = Dimension('arb', 'a.u.', [1])
else: # single force curve
if verbose:
print('Found force curve of size {}'.format(images.shape))
type_suffix = 'ForceCurve'
images = np.atleast_3d(images) # now [Z, chan, 1]
images = images.transpose((1, 2, 0)) # [chan ,1, Z] force curve
# The data generated above varies linearly. Override.
# For now, we'll shove the Z sensor data into the spectroscopic values.
# Find the channel that corresponds to either Z sensor or Raw:
try:
chan_ind = chan_labels.index('ZSnsr')
spec_data = VALUES_DTYPE(images[chan_ind]).squeeze()
except ValueError:
try:
chan_ind = chan_labels.index('Raw')
spec_data = VALUES_DTYPE(images[chan_ind]).squeeze()
except ValueError:
# We don't expect to come here. If we do, spectroscopic values remains as is
spec_data = np.arange(images.shape[2])
pos_desc = Dimension('X', 'm', [1])
spec_desc = Dimension('Z', 'm', spec_data)
# Create measurement group
meas_grp = create_indexed_group(h5_file, grp_name)
# Write file and measurement level parameters
global_parms = generate_dummy_main_parms()
global_parms['data_type'] = 'IgorIBW_' + type_suffix
global_parms['translator'] = 'IgorIBW'
write_simple_attrs(h5_file, global_parms)
write_simple_attrs(meas_grp, parm_dict)
# Create Position and spectroscopic datasets
h5_pos_inds, h5_pos_vals = write_ind_val_dsets(meas_grp, pos_desc, is_spectral=False)
h5_spec_inds, h5_spec_vals = write_ind_val_dsets(meas_grp, spec_desc, is_spectral=True)
# Prepare the list of raw_data datasets
for chan_data, chan_name, chan_unit in zip(images, chan_labels, chan_units):
if verbose:
print('channel', chan_name)
print('unit', chan_unit)
chan_grp = create_indexed_group(meas_grp, 'Channel')
write_main_dataset(chan_grp, np.atleast_2d(chan_data), 'Raw_Data',
chan_name, chan_unit,
None, None,
h5_pos_inds=h5_pos_inds, h5_pos_vals=h5_pos_vals,
h5_spec_inds=h5_spec_inds, h5_spec_vals=h5_spec_vals,
dtype=np.float32)
if verbose:
print('Finished preparing raw datasets')
h5_file.close()
return h5_path
def translate(self, file_path):
"""
The main function that translates the provided file into a .h5 file
Parameters
----------
file_path : String / unicode
Absolute path of any file in the directory
Returns
-------
h5_path : String / unicode
Absolute path of the h5 file
"""
file_path = path.abspath(file_path)
# Figure out the basename of the data:
(basename, parm_paths,
data_paths) = super(GTuneTranslator, self)._parse_file_path(file_path)
(folder_path, unused) = path.split(file_path)
h5_path = path.join(folder_path, basename + '.h5')
if path.exists(h5_path):
remove(h5_path)
# Load parameters from .mat file
matread = loadmat(parm_paths['parm_mat'],
variable_names=[
'AI_wave', 'BE_wave_AO_0', 'BE_wave_AO_1',
'BE_wave_train', 'BE_wave', 'total_cols',
'total_rows'
])
be_wave = np.float32(np.squeeze(matread['BE_wave']))
be_wave_train = np.float32(np.squeeze(matread['BE_wave_train']))
num_cols = int(matread['total_cols'][0][0])
expected_rows = int(matread['total_rows'][0][0])
self.points_per_pixel = len(be_wave)
self.points_per_line = len(be_wave_train)
# Load parameters from .txt file - 'BE_center_frequency_[Hz]', 'IO rate'
is_beps, parm_dict = parmsToDict(parm_paths['parm_txt'])
# Get file byte size:
# For now, assume that bigtime_00 always exists and is the main file
file_size = path.getsize(data_paths[0])
# Calculate actual number of lines since the first few lines may not be saved
self.num_rows = 1.0 * file_size / (4 * self.points_per_pixel *
num_cols)
if self.num_rows % 1:
warn('Error - File has incomplete rows')
return None
else:
self.num_rows = int(self.num_rows)
samp_rate = parm_dict['IO_rate_[Hz]']
ex_freq_nominal = parm_dict['BE_center_frequency_[Hz]']
# method 1 for calculating the correct excitation frequency:
pixel_duration = 1.0 * self.points_per_pixel / samp_rate
num_periods = pixel_duration * ex_freq_nominal
ex_freq_correct = 1 / (pixel_duration / np.floor(num_periods))
# correcting the excitation frequency - will be VERY useful during analysis and filtering
parm_dict['BE_center_frequency_[Hz]'] = ex_freq_correct
# Some very basic information that can help the processing crew
parm_dict['points_per_line'] = self.points_per_line
parm_dict['num_bins'] = self.points_per_pixel
parm_dict['grid_num_rows'] = self.num_rows
parm_dict['data_type'] = 'G_mode_line'
if self.num_rows != expected_rows:
print('Note: {} of {} lines found in data file'.format(
self.num_rows, expected_rows))
# Calculate number of points to read per line:
self.__bytes_per_row__ = int(file_size / self.num_rows)
# First finish writing all global parameters, create the file too:
h5_file = h5py.File(h5_path, 'w')
global_parms = generate_dummy_main_parms()
global_parms['data_type'] = 'G_mode_line'
global_parms['translator'] = 'G_mode_line'
write_simple_attrs(h5_file, global_parms)
# Next create the Measurement and Channel groups and write the appropriate parameters to them
meas_grp = create_indexed_group(h5_file, 'Measurement')
write_simple_attrs(meas_grp, parm_dict)
# Now that the file has been created, go over each raw data file:
"""
We only allocate the space for the main data here.
This does NOT change with each file. The data written to it does.
The auxiliary datasets will not change with each raw data file since
only one excitation waveform is used
"""
pos_desc = Dimension('Y', 'm', np.arange(self.num_rows))
spec_desc = Dimension('Excitation', 'V',
np.tile(VALUES_DTYPE(be_wave), num_cols))
h5_pos_ind, h5_pos_val = write_ind_val_dsets(meas_grp,
pos_desc,
is_spectral=False)
h5_spec_inds, h5_spec_vals = write_ind_val_dsets(meas_grp,
spec_desc,
is_spectral=True)
for f_index in data_paths.keys():
chan_grp = create_indexed_group(meas_grp, 'Channel')
h5_main = write_main_dataset(
chan_grp, (self.num_rows, self.points_per_pixel * num_cols),
'Raw_Data',
'Deflection',
'V',
None,
None,
h5_pos_inds=h5_pos_ind,
h5_pos_vals=h5_pos_val,
h5_spec_inds=h5_spec_inds,
h5_spec_vals=h5_spec_vals,
chunks=(1, self.points_per_pixel),
dtype=np.float16)
# Now transfer scan data in the dat file to the h5 file:
super(GTuneTranslator, self)._read_data(data_paths[f_index],
h5_main)
h5_file.close()
print('G-Tune translation complete!')
return h5_path
def translate(self, file_path, *args, **kwargs):
"""
Translates a given Bruker / Veeco / Nanoscope AFM derived file to HDF5. Currently handles scans, force curves,
and force-distance maps
Note that this translator was written with a single example file for each modality and may be buggy.
Parameters
----------
file_path : str / unicode
path to data file
Returns
-------
h5_path : str / unicode
path to translated HDF5 file
"""
self.file_path = path.abspath(file_path)
self.meta_data, other_parms = self._extract_metadata()
# These files are weirdly named with extensions such as .001
h5_path = file_path.replace('.', '_') + '.h5'
if path.exists(h5_path):
remove(h5_path)
h5_file = h5py.File(h5_path, 'w')
type_suffixes = ['Image', 'Force_Curve', 'Force_Map']
# 0 - stack of scan images
# 1 - single force curve
# 2 - force map
force_count = 0
image_count = 0
for class_name in self.meta_data.keys():
if 'Ciao force image list' in class_name:
force_count += 1
elif 'Ciao image list' in class_name:
image_count += 1
data_type = 0
if force_count > 0:
if image_count > 0:
data_type = 2
else:
data_type = 1
global_parms = generate_dummy_main_parms()
global_parms['data_type'] = 'Bruker_AFM_' + type_suffixes[data_type]
global_parms['translator'] = 'Bruker_AFM'
write_simple_attrs(h5_file, global_parms)
# too many parameters. Making a dummy group just for the parameters.
h5_parms_grp = h5_file.create_group('Parameters')
# We currently have a dictionary of dictionaries. This needs to be flattened
flat_dict = dict()
for class_name, sub_dict in other_parms.items():
for key, val in sub_dict.items():
flat_dict[class_name + '_' + key] = val
write_simple_attrs(h5_parms_grp, flat_dict)
# Create measurement group
h5_meas_grp = create_indexed_group(h5_file, 'Measurement')
# Call the data specific translation function
trans_funcs = [self._translate_image_stack, self._translate_force_curve, self._translate_force_map]
trans_funcs[data_type](h5_meas_grp)
# wrap up and return path
h5_file.close()
return h5_path
def translate(self,
file_path,
show_plots=True,
save_plots=True,
do_histogram=False):
"""
Basic method that translates .dat data file(s) to a single .h5 file
Inputs:
file_path -- Absolute file path for one of the data files.
It is assumed that this file is of the OLD data format.
Outputs:
Nothing
"""
file_path = path.abspath(file_path)
(folder_path, basename) = path.split(file_path)
(basename, path_dict) = self._parse_file_path(file_path)
h5_path = path.join(folder_path, basename + '.h5')
if path.exists(h5_path):
remove(h5_path)
self.h5_file = h5py.File(h5_path, 'w')
isBEPS = True
parm_dict = self.__getParmsFromOldMat(path_dict['old_mat_parms'])
ignored_plt_grps = ['in-field'
] # Here we assume that there is no in-field.
# If in-field data is captured then the translator would have to be modified.
# Technically, we could do away with this if statement, as isBEPS is always true for this translation
if isBEPS:
parm_dict['data_type'] = 'BEPSData'
std_expt = parm_dict[
'VS_mode'] != 'load user defined VS Wave from file'
if not std_expt:
warn(
'This translator does not handle user defined voltage spectroscopy'
)
return
spec_label = getSpectroscopicParmLabel(parm_dict['VS_mode'])
# Check file sizes:
if 'read_real' in path_dict.keys():
real_size = path.getsize(path_dict['read_real'])
imag_size = path.getsize(path_dict['read_imag'])
else:
real_size = path.getsize(path_dict['write_real'])
imag_size = path.getsize(path_dict['write_imag'])
if real_size != imag_size:
raise ValueError(
"Real and imaginary file sizes DON'T match!. Ending")
num_rows = int(parm_dict['grid_num_rows'])
num_cols = int(parm_dict['grid_num_cols'])
num_pix = num_rows * num_cols
tot_bins = real_size / (
num_pix * 4) # Finding bins by simple division of entire datasize
# Check for case where only a single pixel is missing.
check_bins = real_size / ((num_pix - 1) * 4)
if tot_bins % 1 and check_bins % 1:
warn('Aborting! Some parameter appears to have changed in-between')
return
elif not tot_bins % 1:
# Everything's ok
pass
elif not check_bins % 1:
tot_bins = check_bins
warn(
'Warning: A pixel seems to be missing from the data. File will be padded with zeros.'
)
tot_bins = int(tot_bins)
(bin_inds, bin_freqs, bin_FFT, ex_wfm,
dc_amp_vec) = self.__readOldMatBEvecs(path_dict['old_mat_parms'])
"""
Because this is the old data format and there is a discrepancy in the number of bins (they seem to be 2 less
than the actual number), we need to re-calculate it based on the available data. This is done below.
"""
band_width = parm_dict['BE_band_width_[Hz]'] * (
0.5 - parm_dict['BE_band_edge_trim'])
st_f = parm_dict['BE_center_frequency_[Hz]'] - band_width
en_f = parm_dict['BE_center_frequency_[Hz]'] + band_width
bin_freqs = np.linspace(st_f, en_f, len(bin_inds), dtype=np.float32)
# Forcing standardized datatypes:
bin_inds = np.int32(bin_inds)
bin_freqs = np.float32(bin_freqs)
bin_FFT = np.complex64(bin_FFT)
ex_wfm = np.float32(ex_wfm)
self.FFT_BE_wave = bin_FFT
(UDVS_labs, UDVS_units, UDVS_mat) = self.__buildUDVSTable(parm_dict)
# Remove the unused plot group columns before proceeding:
(UDVS_mat, UDVS_labs, UDVS_units) = trimUDVS(UDVS_mat, UDVS_labs,
UDVS_units,
ignored_plt_grps)
spec_inds = np.zeros(shape=(2, tot_bins), dtype=INDICES_DTYPE)
# Will assume that all excitation waveforms have same number of bins
# Here, the denominator is 2 because only out of field measruements. For IF + OF, should be 1
num_actual_udvs_steps = UDVS_mat.shape[0] / 2
bins_per_step = tot_bins / num_actual_udvs_steps
# Some more checks
if bins_per_step % 1:
warn('Non integer number of bins per step!')
return
else:
bins_per_step = int(bins_per_step)
num_actual_udvs_steps = int(num_actual_udvs_steps)
stind = 0
for step_index in range(UDVS_mat.shape[0]):
if UDVS_mat[step_index, 2] < 1E-3: # invalid AC amplitude
continue # skip
spec_inds[0, stind:stind + bins_per_step] = np.arange(
bins_per_step, dtype=INDICES_DTYPE) # Bin step
spec_inds[1, stind:stind + bins_per_step] = step_index * np.ones(
bins_per_step, dtype=INDICES_DTYPE) # UDVS step
stind += bins_per_step
del stind, step_index
# Some very basic information that can help the processing / analysis crew
parm_dict['num_bins'] = tot_bins
parm_dict['num_pix'] = num_pix
parm_dict['num_udvs_steps'] = num_actual_udvs_steps
global_parms = generate_dummy_main_parms()
global_parms['grid_size_x'] = parm_dict['grid_num_cols']
global_parms['grid_size_y'] = parm_dict['grid_num_rows']
global_parms['experiment_date'] = parm_dict['File_date_and_time']
# assuming that the experiment was completed:
global_parms['current_position_x'] = parm_dict['grid_num_cols'] - 1
global_parms['current_position_y'] = parm_dict['grid_num_rows'] - 1
global_parms['data_type'] = parm_dict[
'data_type'] # self.__class__.__name__
global_parms['translator'] = 'ODF'
write_simple_attrs(self.h5_file, global_parms)
# Create Measurement and Channel groups
meas_grp = create_indexed_group(self.h5_file, 'Measurement')
write_simple_attrs(meas_grp, parm_dict)
chan_grp = create_indexed_group(meas_grp, 'Channel')
chan_grp.attrs['Channel_Input'] = parm_dict['IO_Analog_Input_1']
# Create Auxilliary Datasets
h5_ex_wfm = chan_grp.create_dataset('Excitation_Waveform', data=ex_wfm)
udvs_slices = dict()
for col_ind, col_name in enumerate(UDVS_labs):
udvs_slices[col_name] = (slice(None), slice(col_ind, col_ind + 1))
h5_UDVS = chan_grp.create_dataset('UDVS',
data=UDVS_mat,
dtype=np.float32)
write_simple_attrs(h5_UDVS, {'labels': UDVS_labs, 'units': UDVS_units})
h5_bin_steps = chan_grp.create_dataset('Bin_Steps',
data=np.arange(bins_per_step,
dtype=np.uint32),
dtype=np.uint32)
# Need to add the Bin Waveform type - infer from UDVS
exec_bin_vec = self.signal_type * np.ones(len(bin_inds),
dtype=np.int32)
h5_wfm_typ = chan_grp.create_dataset('Bin_Wfm_Type',
data=exec_bin_vec,
dtype=np.int32)
h5_bin_inds = chan_grp.create_dataset('Bin_Indices',
data=bin_inds,
dtype=np.uint32)
h5_bin_freq = chan_grp.create_dataset('Bin_Frequencies',
data=bin_freqs,
dtype=np.float32)
h5_bin_FFT = chan_grp.create_dataset('Bin_FFT',
data=bin_FFT,
dtype=np.complex64)
# Noise floor should be of shape: (udvs_steps x 3 x positions)
h5_noise_floor = chan_grp.create_dataset(
'Noise_Floor',
shape=(num_pix, num_actual_udvs_steps),
dtype=nf32,
chunks=(1, num_actual_udvs_steps))
"""
ONLY ALLOCATING SPACE FOR MAIN DATA HERE!
Chunk by each UDVS step - this makes it easy / quick to:
1. read data for a single UDVS step from all pixels
2. read an entire / multiple pixels at a time
The only problem is that a typical UDVS step containing 50 steps occupies only 400 bytes.
This is smaller than the recommended chunk sizes of 10,000 - 999,999 bytes
meaning that the metadata would be very substantial.
This assumption is fine since we almost do not handle any user defined cases
"""
"""
New Method for chunking the Main_Data dataset. Chunking is now done in N-by-N squares of UDVS steps by pixels.
N is determined dinamically based on the dimensions of the dataset. Currently it is set such that individual
chunks are less than 10kB in size.
Chris Smith -- [email protected]
"""
pos_dims = [
Dimension('X', 'nm', num_cols),
Dimension('Y', 'nm', num_rows)
]
# Create Spectroscopic Values and Spectroscopic Values Labels datasets
spec_vals, spec_inds, spec_vals_labs, spec_vals_units, spec_vals_names = createSpecVals(
UDVS_mat, spec_inds, bin_freqs, exec_bin_vec, parm_dict, UDVS_labs,
UDVS_units)
spec_dims = list()
for row_ind, row_name in enumerate(spec_vals_labs):
spec_dims.append(
Dimension(row_name, spec_vals_units[row_ind],
spec_vals[row_ind]))
pixel_chunking = maxReadPixels(10240, num_pix * num_actual_udvs_steps,
bins_per_step,
np.dtype('complex64').itemsize)
chunking = np.floor(np.sqrt(pixel_chunking))
chunking = max(1, chunking)
chunking = min(num_actual_udvs_steps, num_pix, chunking)
self.h5_main = write_main_dataset(chan_grp, (num_pix, tot_bins),
'Raw_Data',
'Piezoresponse',
'V',
pos_dims,
spec_dims,
dtype=np.complex64,
chunks=(chunking,
chunking * bins_per_step),
compression='gzip')
self.mean_resp = np.zeros(shape=(self.ds_main.shape[1]),
dtype=np.complex64)
self.max_resp = np.zeros(shape=(self.ds_main.shape[0]),
dtype=np.float32)
self.min_resp = np.zeros(shape=(self.ds_main.shape[0]),
dtype=np.float32)
# Now read the raw data files:
self._read_data(path_dict['read_real'], path_dict['read_imag'],
parm_dict)
self.h5_file.flush()
generatePlotGroups(self.ds_main,
self.mean_resp,
folder_path,
basename,
self.max_resp,
self.min_resp,
max_mem_mb=self.max_ram,
spec_label=spec_label,
show_plots=show_plots,
save_plots=save_plots,
do_histogram=do_histogram)
self.h5_file.close()
return h5_path
def translate(self, parm_path):
"""
Basic method that translates .mat data files to a single .h5 file
Parameters
------------
parm_path : string / unicode
Absolute file path of the parameters .mat file.
Returns
----------
h5_path : string / unicode
Absolute path of the translated h5 file
"""
self.parm_path = path.abspath(parm_path)
(folder_path, file_name) = path.split(parm_path)
(file_name, base_name) = path.split(folder_path)
h5_path = path.join(folder_path, base_name + '.h5')
# Read parameters
parm_dict = readGmodeParms(parm_path)
# Add the w^2 specific parameters to this list
parm_data = loadmat(parm_path, squeeze_me=True, struct_as_record=True)
#freq_sweep_parms = parm_data['freqSweepParms']
#parm_dict['freq_sweep_delay'] = np.float(freq_sweep_parms['delay'].item())
gen_sig = parm_data['genSig']
#parm_dict['wfm_fix_d_fast'] = np.int32(gen_sig['restrictT'].item())
#freq_array = np.float32(parm_data['freqArray'])
# prepare and write spectroscopic values
samp_rate = parm_dict['IO_down_samp_rate_[Hz]']
num_bins = int(parm_dict['wfm_n_cycles'] * parm_dict['wfm_p_slow'] * samp_rate)
w_vec = np.arange(-0.5 * samp_rate, 0.5 * samp_rate, np.float32(samp_rate / num_bins))
# There is most likely a more elegant solution to this but I don't have the time... Maybe np.meshgrid
spec_val_mat = np.zeros((len(freq_array) * num_bins, 2), dtype=VALUES_DTYPE)
spec_val_mat[:, 0] = np.tile(w_vec, len(freq_array))
spec_val_mat[:, 1] = np.repeat(freq_array, num_bins)
spec_ind_mat = np.zeros((2, len(freq_array) * num_bins), dtype=np.int32)
spec_ind_mat[0, :] = np.tile(np.arange(num_bins), len(freq_array))
spec_ind_mat[1, :] = np.repeat(np.arange(len(freq_array)), num_bins)
num_rows = parm_dict['grid_num_rows']
num_cols = parm_dict['grid_num_cols']
parm_dict['data_type'] = 'GVS'
num_pix = num_rows * num_cols
global_parms = generate_dummy_main_parms()
global_parms['grid_size_x'] = parm_dict['grid_num_cols']
global_parms['grid_size_y'] = parm_dict['grid_num_rows']
# assuming that the experiment was completed:
global_parms['current_position_x'] = parm_dict['grid_num_cols'] - 1
global_parms['current_position_y'] = parm_dict['grid_num_rows'] - 1
global_parms['data_type'] = parm_dict['data_type'] # self.__class__.__name__
global_parms['translator'] = 'GVS'
# Now start creating datasets and populating:
if path.exists(h5_path):
remove(h5_path)
h5_f = h5py.File(h5_path, 'w')
write_simple_attrs(h5_f, global_parms)
meas_grp = create_indexed_group(h5_f, 'Measurement')
chan_grp = create_indexed_group(meas_grp, 'Channel')
write_simple_attrs(chan_grp, parm_dict)
pos_dims = [Dimension('X', 'nm', num_rows),
Dimension('Y', 'nm', num_cols)]
spec_dims = [Dimension('Response Bin', 'a.u.', num_bins),
Dimension('Excitation Frequency ', 'Hz', len(freq_array))]
# Minimize file size to the extent possible.
# DAQs are rated at 16 bit so float16 should be most appropriate.
# For some reason, compression is more effective on time series data
h5_main = write_main_dataset(chan_grp, (num_pix, num_bins), 'Raw_Data',
'Deflection', 'V',
pos_dims, spec_dims,
chunks=(1, num_bins), dtype=np.float32)
h5_ex_freqs = chan_grp.create_dataset('Excitation_Frequencies', freq_array)
h5_bin_freq = chan_grp.create_dataset('Bin_Frequencies', w_vec)
# Now doing link_h5_objects_as_attrs:
link_h5_objects_as_attrs(h5_main, [h5_ex_freqs, h5_bin_freq])
# Now read the raw data files:
pos_ind = 0
for row_ind in range(1, num_rows + 1):
for col_ind in range(1, num_cols + 1):
file_path = path.join(folder_path, 'fSweep_r' + str(row_ind) + '_c' + str(col_ind) + '.mat')
print('Working on row {} col {}'.format(row_ind, col_ind))
if path.exists(file_path):
# Load data file
pix_data = loadmat(file_path, squeeze_me=True)
pix_mat = pix_data['AI_mat']
# Take the inverse FFT on 2nd dimension
pix_mat = np.fft.ifft(np.fft.ifftshift(pix_mat, axes=1), axis=1)
# Verified with Matlab - no conjugate required here.
pix_vec = pix_mat.transpose().reshape(pix_mat.size)
h5_main[pos_ind, :] = np.float32(pix_vec)
h5_f.flush() # flush from memory!
else:
print('File not found for: row {} col {}'.format(row_ind, col_ind))
pos_ind += 1
if (100.0 * pos_ind / num_pix) % 10 == 0:
print('completed translating {} %'.format(int(100 * pos_ind / num_pix)))
h5_f.close()
return h5_path
