Generate sensitivity tuning curve fails

[akandrew332]

Installed toffy successfully on CAC, however no rosetta_matrices folder created. Additionally, generating sensitivity curve fails. Identified good sweep for generating tuning curve. What exactly is supposed to be copied to the normalization curve folder? Instructions state to copy all FOVs into that folder. Each voltage in the sweep is a different folder itself. Tried copying all the different folders for moly scans at different voltages for the sweep into the normalization curve folder. This failed. Are only specific files supposed to placed in the normalization curve folder? Thanks!

[ngreenwald]

Yes, you should be able to just directly copy each run folder into the overall normalization folder.

i.e.

normalization_curve
│ 
├── 20220901_2200V
│    ├── fov-1-scan-1.bin
│    ├── fov-1-scan-1.json
├── 20220901_2225V
│    ├── fov-1-scan-1.bin
│    ├── fov-1-scan-1.json
├── 20220901_2250V
│    ├── fov-1-scan-1.bin
│    ├── fov-1-scan-1.json

Good catch about the rosetta matrices folder, we'll add that into the notebook

[camisowers]

Thank you!

[akandrew332]

No problem! By the way, those wider range plots might have some value... Here is our current detector:

[akandrew332]

Here's a previous faulty one:

[akandrew332]

Should we be doing normalization curves every time the detector is replaced?

[ngreenwald]

The outliers between 10k and 20k (on the x-axis) are what makes the plot look different. If you exclude those, you can imagine that all curves have an elbow point around 8k (on the x-axis), which is what we're looking for, that the shape of the curve is the same. We don't know for sure, but so far haven't seen significant differences between detectors.