Week of January 16-20:
- There were three Auxtel runs that happened on 17th, 18th, and 19th.
- We had unexpected issues:
- Camera was facing somewhere other than the pegboard.
- The camera was out of focus.
- We had to ask for assistance from observing specialists.
- The records show that we had a 5 hour run on the night of January 17.
- For the night of January 18th and 19th we could not take data because Camera gave a "no more shots error" and was unaccessible. Restarting it did not fix the problem.
- Positive updates from the week:
- We can take data in plain daylight.
- We solved the problem of overriding in the computer. When the file names reset, it won't be a problem in the computer anymore. It may still be a problem in camera. The camera storage seems like it has images we can't delete.
- Currently we have data from the night of the January 17. The path forward is to make a preliminary analysis on this data and see whether a paper could be written from it.
- Steps of the analysis:
- Take a look at the psfs in a 15 minute interval and see if they are consistent.
- If they are consistent do a differential image motion analysis where the data is convolved with a large Gaussian.
- Check the evolution of saturation values on the 15 minute interval and see if they can be correlated with wind gusts and temperature fluctuation.
Week of January 23-27:
- The data is currently in the cluster and I have been trying to run a Jupyter notebook in the cluster in which we use the environment
(/n/holyscratch01/stubbs_lab/Lab/python/LSST_DESC). This environment has rawpy and sep (source extractor python) in it.
- When I try to run an ipython notebook with this environment, it works.
- However, I can't run a jupyter notebook. I followed the instructions both on Harvard Research Computing Notes and https://docs.rc.fas.harvard.edu/kb/tag/jupyter-notebook/.
- After a lot of debugging I realized what the problem was. When a user wants to access a shared environment of a lab, they add the path of the environment in their ~./.conda.envrironments.txt. In our case it is /n/holyscratch01/stubbs_lab/Lab/python/LSST_DESC
- I have tried doing this, with nano, vim and simple bash scripting. Every time I start a jupyter session. The environment location gets deleted from my ~./.conda.envrironments.txt
An example command:
Code Block echo "/n/holyscratch01/stubbs_lab/Lab/python/LSST_DESC" >> ~/.conda/environments.txt- After running this and looking at the contents of the file with either less, more or cat:
It looks like the statement is written. However as soon as my vdi jupyter session gets activated, it gets deleted like it is shown on the second line. I have been stuck on this part.