occheung
/
artiq
forked from M-Labs/artiq
1
0
Fork 0
Code Pull Requests Activity

examples: reconstruct scans when analyze()ing HDF5 files

This commit is contained in:
Robert Jördens 2016-05-12 12:59:36 +02:00
parent cd1304547d
commit 5d58258bf2
1 changed files with 10 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
artiq/examples/master/repository/flopping_f_simulation.py
View File

@ -57,8 +57,17 @@ class FloppingF(EnvExperiment):
def analyze(self):
# Use get_dataset so that analyze can be run stand-alone.
frequency = self.get_dataset("flopping_f_frequency")
brightness = self.get_dataset("flopping_f_brightness")
try:
frequency = self.get_dataset("flopping_f_frequency")
except KeyError:
# Since flopping_f_frequency is not saved, it is missing if
# analyze() is run on HDF5 data. But assuming that the arguments
# have been loaded from that same HDF5 file, we can reconstruct it.
frequency = np.fromiter(self.frequency_scan, np.float)
assert frequency.shape == brightness.shape
self.set_dataset("flopping_f_frequency", frequency,
broadcast=True, save=False)
popt, pcov = curve_fit(model_numpy,
frequency, brightness,
p0=[self.get_dataset("flopping_freq", 1500.0)])