Create Rfrac plot when bias steps are taken and run_analysis=True

[msilvafe]

I think this addresses issue #343 so that we will have Rfrac plots made regularly viewable via Caleb's HTML viewer.

[jlashner]

A couple of small things, we want this to only plot when take_bias_steps is run directly, and not through take_bgmap. I think we should add a plot_rfrac=True kwarg to take_bias_steps that gets set to False in take_bgmap. Also I believe the plot_Rfrac function just returns the figure but doesn't save it.

[jlashner]

Thanks max, this looks good to me! @dpdutcher or @samdayweiss any chance you have a system where we can quickly test this? Otherwise I'm fine with just merging.

[dpdutcher]

Not at the moment, but I should late next week

[samdayweiss]

We had the SMuRFs off for the last few days while we did some work on their mount but I’m gonna be doing some detector work Monday and Tuesday so we can test then? Let me know, Sam Get Outlook for iOS<https://aka.ms/o0ukef>

…

________________________________ From: dpdutcher ***@***.***> Sent: Friday, May 26, 2023 4:02:34 PM To: simonsobs/sodetlib ***@***.***> Cc: Samuel F. Day-Weiss ***@***.***>; Mention ***@***.***> Subject: Re: [simonsobs/sodetlib] Create Rfrac plot when bias steps are taken and run_analysis=True (PR #345) Not at the moment, but I should late next week — Reply to this email directly, view it on GitHub<#345 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ATJIAG776HZIZKNDZCITTETXIED5VANCNFSM6AAAAAAYOHQ3CM>. You are receiving this because you were mentioned.Message ID: ***@***.***>

[samdayweiss]

Testing this on SATP3 now

[samdayweiss]

Running into some errors when running tage_bgmap and other functions that call set_current_mode. Traceback for bias_to_rfrac:

`AttributeError Traceback (most recent call last)
File :1

File /sodetlib/sodetlib/operations/bias_dets.py:198, in bias_to_rfrac(S, cfg, rfrac, bias_groups, iva, overbias, Rn_range, math_only)
195 S.log(f"BG {bg}: {biases[bg]:.2f}")
197 if overbias:
--> 198 sdl.overbias_dets(S, cfg, bias_groups=bias_groups)
200 S.set_tes_bias_bipolar_array(biases)
202 return biases

File /sodetlib/sodetlib/util.py:851, in overbias_dets(S, cfg, bias_groups, biases, cool_wait, high_current_mode)
848 high_current_mode = [high_current_mode for _ in range(12)]
850 S.log("Overbiasing Detectors")
--> 851 set_current_mode(S, bias_groups, 1)
852 for bg in bias_groups:
853 S.set_tes_bias_bipolar(bg, cfg.dev.bias_groups[bg]['overbias_voltage'])

File /sodetlib/sodetlib/util.py:682, in set_current_mode(S, bgs, mode, const_current)
677 # It takes longer for DC voltages to settle than it does to toggle the
678 # high-current relay, so we can set them at the same time when switchign
679 # to hcm, but when switching to lcm we need a sleep statement to prevent
680 # dets from latching.
681 if mode:
--> 682 epics.caput_many([S.C.writepv, dac_data_reg], [relay_data, dac_data],
683 wait=True)
684 else:
685 S._caput(dac_data_reg, dac_data)

File /usr/local/lib/python3.8/dist-packages/epics/init.py:362, in caput_many(pvlist, values, wait, connection_timeout, put_timeout)
360 raise ValueError("List of PV names must be equal to list of values.")
361 out = []
--> 362 pvs = [PV(name, auto_monitor=False, connection_timeout=connection_timeout) for name in pvlist]
363 conns = [p.connected for p in pvs]
364 wait_all = (wait == 'all')

File /usr/local/lib/python3.8/dist-packages/epics/init.py:362, in (.0)
360 raise ValueError("List of PV names must be equal to list of values.")
361 out = []
--> 362 pvs = [PV(name, auto_monitor=False, connection_timeout=connection_timeout) for name in pvlist]
363 conns = [p.connected for p in pvs]
364 wait_all = (wait == 'all')

File /usr/local/lib/python3.8/dist-packages/epics/pv.py:205, in PV.init(self, pvname, callback, form, verbose, auto_monitor, count, connection_callback, connection_timeout, access_callback)
199 def init(self, pvname, callback=None, form='time',
200 verbose=False, auto_monitor=None, count= None,
201 connection_callback=None,
202 connection_timeout=None,
203 access_callback=None):
--> 205 self.pvname = pvname.strip()
206 self.form = form.lower()
207 self.verbose = verbose

AttributeError: 'PV' object has no attribute 'strip'`

Traceback for take_bgmap:

`---------------------------------------------------------------------------
AttributeError Traceback (most recent call last)
File /sodetlib/sodetlib/operations/bias_steps.py:1175, in take_bias_steps(S, cfg, bgs, step_voltage, step_duration, nsteps, high_current_mode, hcm_wait_time, run_analysis, analysis_kwargs, dacs, use_waveform, channel_mask, g3_tag, plot_rfrac, show_plots)
1174 step_voltage /= S.high_low_current_ratio
-> 1175 sdl.set_current_mode(S, bgs, 1)
1176 S.log(f"Waiting {hcm_wait_time} sec after switching to hcm")

File /sodetlib/sodetlib/util.py:682, in set_current_mode(S, bgs, mode, const_current)
681 if mode:
--> 682 epics.caput_many([S.C.writepv, dac_data_reg], [relay_data, dac_data],
683 wait=True)
684 else:

File /usr/local/lib/python3.8/dist-packages/epics/init.py:362, in caput_many(pvlist, values, wait, connection_timeout, put_timeout)
361 out = []
--> 362 pvs = [PV(name, auto_monitor=False, connection_timeout=connection_timeout) for name in pvlist]
363 conns = [p.connected for p in pvs]

File /usr/local/lib/python3.8/dist-packages/epics/init.py:362, in (.0)
361 out = []
--> 362 pvs = [PV(name, auto_monitor=False, connection_timeout=connection_timeout) for name in pvlist]
363 conns = [p.connected for p in pvs]

File /usr/local/lib/python3.8/dist-packages/epics/pv.py:205, in PV.init(self, pvname, callback, form, verbose, auto_monitor, count, connection_callback, connection_timeout, access_callback)
199 def init(self, pvname, callback=None, form='time',
200 verbose=False, auto_monitor=None, count= None,
201 connection_callback=None,
202 connection_timeout=None,
203 access_callback=None):
--> 205 self.pvname = pvname.strip()
206 self.form = form.lower()

AttributeError: 'PV' object has no attribute 'strip'

During handling of the above exception, another exception occurred:

TypeError Traceback (most recent call last)
Cell In[6], line 1
----> 1 bsa = bias_steps.take_bgmap(S, cfg)

File /usr/local/src/pysmurf/python/pysmurf/client/util/pub.py:50, in set_action..dec..wrapper(S, pub_action, *args, **kwargs)
46 S.pub._action = func.name
48 S.pub._action_ts = S.get_timestamp()
---> 50 rv = func(S, *args, **kwargs)
52 finally:
53 if is_top:

File /sodetlib/sodetlib/operations/bias_steps.py:1065, in take_bgmap(S, cfg, bgs, dc_voltage, step_voltage, step_duration, nsteps, high_current_mode, hcm_wait_time, analysis_kwargs, dacs, use_waveform, show_plots, g3_tag, plot_rfrac)
1060 _analysis_kwargs = {
1061 'assignment_thresh': 0.3,
1062 'create_bg_map': True, 'save_bg_map': True
1063 }
1064 _analysis_kwargs.update(analysis_kwargs)
-> 1065 bsa = take_bias_steps(
1066 S, cfg, bgs, step_voltage=step_voltage, step_duration=step_duration,
1067 nsteps=nsteps, high_current_mode=high_current_mode,
1068 hcm_wait_time=hcm_wait_time, run_analysis=True, dacs=dacs,
1069 use_waveform=use_waveform, g3_tag=g3_tag, plot_rfrac=plot_rfrac,
1070 show_plots=show_plots, analysis_kwargs=_analysis_kwargs
1071 )
1073 if hasattr(bsa, 'bgmap'):
1074 fig, ax = plot_bg_assignment(bsa)

File /usr/local/src/pysmurf/python/pysmurf/client/util/pub.py:50, in set_action..dec..wrapper(S, pub_action, *args, **kwargs)
46 S.pub._action = func.name
48 S.pub._action_ts = S.get_timestamp()
---> 50 rv = func(S, *args, **kwargs)
52 finally:
53 if is_top:

File /sodetlib/sodetlib/operations/bias_steps.py:1205, in take_bias_steps(S, cfg, bgs, step_voltage, step_duration, nsteps, high_current_mode, hcm_wait_time, run_analysis, analysis_kwargs, dacs, use_waveform, channel_mask, g3_tag, plot_rfrac, show_plots)
1202 sdl.stream_g3_off(S)
1204 # Restores current mode to initial values
-> 1205 sdl.set_current_mode(S, np.where(init_current_mode == 0)[0], 0)
1206 sdl.set_current_mode(S, np.where(init_current_mode == 1)[0], 1)
1208 S.set_downsample_factor(initial_ds_factor)

File /sodetlib/sodetlib/util.py:687, in set_current_mode(S, bgs, mode, const_current)
685 S._caput(dac_data_reg, dac_data)
686 time.sleep(0.04)
--> 687 S._caput(S.C.writepv, relay_data)
689 time.sleep(0.1)

File /usr/local/src/pysmurf/python/pysmurf/client/command/smurf_command.py:104, in SmurfCommandMixin._caput(self, pvname, val, write_log, execute, wait_before, wait_after, wait_done, log_level, enable_poll, disable_poll, new_epics_root, **kwargs)
100 self.log(log_str, log_level)
102 if execute and not self.offline:
103 #epics.caput(pvname, val, wait=wait_done, **kwargs)
--> 104 if pvname not in self._pv_cache.keys():
105 self._pv_cache[pvname] = epics.PV(pvname)
106 self._pv_cache[pvname].put(val,
107 wait=wait_done,
108 **kwargs)

TypeError: unhashable type: 'PV'`

[samdayweiss]

Test complete after fixes from Jack and the plots show up nicely on Teleview, I think this can be merged!