[AGIPD][FIX] Fix overflow bug on plotting baseline shift if results casted to float16

Description

This fixes overflow issue which happens in case results cast to float16

Traceback

ValueError                                Traceback (most recent call last)
Cell In[41], line 4
2 corrected_ave = np.nansum(corrected, axis=(2, 3))
3 plt.scatter(corrected_ave.flatten()/10**6, blshift.flatten(), s=0.9)
----> 4 plt.xlim(np.nanpercentile(corrected_ave/10**6, [2, 98])) 5 plt.grid()
6 plt.xlabel('Illuminated corrected [MADU] ')
File ~/deployments/development/git.xfel.eu/detectors/pycalibration/current/.venv/lib/ 􏰁→python3.11/site-packages/matplotlib/pyplot.py:1961, in xlim(*args, **kwargs)
1959 if not args and not kwargs:
1960 return ax.get_xlim()
-> 1961 ret = ax.set_xlim(*args, **kwargs)
1962 return ret
File ~/deployments/development/git.xfel.eu/detectors/pycalibration/current/.venv/lib/ 􏰁→python3.11/site-packages/matplotlib/axes/_base.py:3664, in _AxesBase.set_xlim(self,␣
􏰁→left, 3662 3663 -> 3664
right, emit, auto, xmin, xmax)
raise TypeError("Cannot pass both 'right' and 'xmax'")
right = xmax
return self.xaxis._set_lim(left, right, emit=emit, auto=auto)
File ~/deployments/development/git.xfel.eu/detectors/pycalibration/current/.venv/lib/ 􏰁→python3.11/site-packages/matplotlib/axis.py:1210, in Axis._set_lim(self, v0, v1,␣ 􏰁→emit, auto)
1207 name = self._get_axis_name()
1209 self.axes._process_unit_info([(name, (v0, v1))], convert=False) -> 1210 v0 = self.axes._validate_converted_limits(v0, self.convert_units) 1211 v1 = self.axes._validate_converted_limits(v1, self.convert_units)
1213 if v0 is None or v1 is None:
1214 # Axes init calls set_xlim(0, 1) before get_xlim() can be called,
1215 # so only grab the limits if we really need them.
File ~/deployments/development/git.xfel.eu/detectors/pycalibration/current/.venv/lib/ 􏰁→python3.11/site-packages/matplotlib/axes/_base.py:3585, in _AxesBase._validate_ 􏰁→converted_limits(self, limit, convert)
3582 converted_limit = converted_limit.squeeze() 3583 if (isinstance(converted_limit, Real)
3584 and not np.isfinite(converted_limit)):
-> 3585 raise ValueError("Axis limits cannot be NaN or Inf") 3586 return converted_limit
ValueError: Axis limits cannot be NaN or Inf

How Has This Been Tested?

hotfixed in deployment

Relevant Documents (optional)

SPB_DET_AGIPD1M-1_correct_006880_r0359_241025_112650_413461.pdf

Types of changes

• Bug fix (non-breaking change which fixes an issue)

Reviewers

@ahmedk @schmidtp