Plots in basic_walkthrough.ipynb should be updated to reflect timeseries units, line-of-sight convention

[Shima-shoki]

Hello, I've been testing the basic walkthrough codes on Google Colab. I could run the code, but it looks like the result is not consistent with the example on the page. The figure shows the PS output in "radians", and the value ranges from around -30 to 30. However, my result shows the value around -0.1 ~ 0.1 as shown in the first figure. The general appearance is quite similar to the example output except the range and sign of the values. My guess is that the new result is in the unit of "m", rather than radian. Is this due to a change in the internal software, or do I have any option to change the output unit, or am I doing something wrong? Any help is very much appreciated!

The .ipynb file, dolphin.log, and dolphin_config.yaml are available from this zip:
dolphin_walkthough_codes.zip

Figure 1. My output.

Figure 2. The reference output in the "basic walkthrough" page.

[scottstanie]

Ah you're right, thanks for spotting that.
I hadn't let it run to the end and checked on the results in awhile:

• Units are now meters, but more importantly
• LOS convention for time timeseries/ outputs has flipped (to match isce2, mintpy) so that positive values indicate motion toward the satellite (instead a negative unwrapped phase value indicating a decrease in line of sight distance i.e. an uplift)

I need to change the point-picking line at the end from row, col = np.unravel_index(np.argmin(last_displacement), last_displacement.shape) to use np.argmax.
But if I'm updating, I might as well add a few more explanatory plots showing the temporal coherence, etc.

[Shima-shoki]

@scottstanie
Thank you! Well noted with changes in the unit, and the LOS convention :)