def plot_z_fitpar_from_average(fig,
fit_par,
img_id,
channel,
led_arrays,
window_size=51):
# make led_arrays a tuple
if type(led_arrays) == int:
led_arrays = (led_arrays, )
fit_parameters = calc.read_hdf(channel)
fit_parameters = calc.include_column_if_nonexistent(
fit_parameters, fit_par, channel)
cropped_parameters = fit_parameters.loc[(
float(img_id -
(window_size - 1) // 2), ):(float(img_id +
(window_size - 1) // 2), )][[
'line', fit_par, 'height'
]]
mean = cropped_parameters.mean(axis=0, level='led_id')
ax = fig.gca(xlabel=fit_par, ylabel='height/m')
for line in led_arrays:
plot, = ax.plot(np.array(mean[mean['line'] == line][fit_par]),
np.array(mean[mean['line'] == line]['height']))
plot.set_label(f'LED_Array{line}, C{channel}')
ax.legend()
plt.title(
f'Plot of averaged fit parameter {fit_par} over time against the height.\n'
f'Image: {img_id}, window_size: {window_size}')
def _calc_t_fitpar_plot_info(led_id, fit_par, channel, image_id_start, image_id_finish):
times = led.load_file(".{}analysis{}image_infos_analysis.csv".format(sep, sep), delim=',', dtype=str)
times = pd.DataFrame(times[:, [0, 3]], columns=['img_id', 'experiment_time'], dtype=np.float64)
times.set_index('img_id', inplace=True)
fit_parameters = calc.read_hdf(channel)
fit_parameters = calc.include_column_if_nonexistent(fit_parameters, fit_par, channel)
idx = pd.IndexSlice
fit_parameters = fit_parameters.loc[idx[:, led_id], fit_par]
fit_parameters.reset_index(drop=True, level=1, inplace=True)
plot_info = pd.concat([times, fit_parameters], axis=1, sort=False)
plot_info = plot_info.loc[idx[image_id_start:image_id_finish+1]]
return plot_info
def plot_z_fitpar(fig: plt.figure, fit_par: str, img_id: int, channel: int,
led_arrays: Union[Tuple[int, ...], int]) -> None:
"""plots the height of a LED array against one fit parameter"""
# make led_arrays a tuple
if type(led_arrays) == int:
led_arrays = (led_arrays,)
fit_parameters = calc.read_hdf(channel)
fit_parameters = calc.include_column_if_nonexistent(fit_parameters, fit_par, channel)
fit_parameters = fit_parameters.loc[img_id, :]
ax = fig.gca(xlabel=fit_par, ylabel='height/m')
for line in led_arrays:
plot, = ax.plot(np.array(fit_parameters[fit_parameters['line'] == line][fit_par]),
np.array(fit_parameters[fit_parameters['line'] == line]['height']))
plot.set_label(f'LED_Array{line}, C{channel}')
ax.legend()
plt.title(f'Plot of fit parameter {fit_par} against the height.\n'
f'Image: {img_id}')
def load_model(img_id, led_id, channel, window_radius=10):
fit_parameters = calc.read_hdf(channel)
model_params = np.array(fit_parameters.loc[img_id, led_id])[1:9]
pix_pos = get_led_pos(led_id)
model_params[0:2] = model_params[0:2] - pix_pos + window_radius
return model_params
def load_img_data(self) -> None:
img_data = read_hdf(self.experiment.channel, path=self.experiment.path)
img_data_cropped = img_data[['line', self.reference_property]]
self.calculated_img_data = img_data_cropped[img_data_cropped['line'] ==
self.experiment.led_array]