AIO for September 9, 2014
Last week's accomplishments
Objectives from last week:
- Fix the permissions issues for offline backups for the HMC website. -- Done, the backups are now collecting on my server twice a month. See https://github.com/csu-hmc/hmc.csuohio.edu#backup for details of the implementation.
- Implement the 50X faster constraints Jacobian evaluation for the pendulum id problem. -- Done, didn't quite get 50x increase, but iterations for 6000 nodes are down to 1 second an iteration. Now IPOPT is the slow part and the function evaluations are in general faster. There is an interesting thread at http://list.coin-or.org/pipermail/ipopt/2014-September/003817.html discussing how to improve the IPOPT speed.
- Write the draft of the raw data section in the perturbed data paper. -- Done, see https://github.com/csu-hmc/perturbed-data-paper/blob/master/paper.tex.
- Add more detail to direct collocation id outline. -- Done, https://github.com/csu-hmc/inverted-pendulum-sys-id-paper.
- Write an introduction to the direct collocation id paper. -- Done, https://github.com/csu-hmc/inverted-pendulum-sys-id-paper/blob/master/paper.tex.
- Fix all remaining issues on the yeadon paper and submit to F1000Research. -- Done, see https://github.com/chrisdembia/python-yeadon-paper/pull/30.
- Create a figshare item for the raw bicycle data. -- Done, I've created a fileset for each the trial data and the calibration data.
- Touch base with Luke on visiting scholar idea. -- Done, sent him an email, but no response.
- Think of the exercises I will use for teaching git/github in the open source software night. -- Done, developed the teaching plans more in the google doc.
- Advertise for open source software night. -- Done, sent emails to ME profs and the CS department plus posted flyers on the ME floor in Stillwell.
Other things I did:
- Submitted a patch to SymPy about the Fortran printers: https://github.com/sympy/sympy/pull/7968
- Helped Ton with Git/Github.
- Sat in on Ton's class.
- Submitted a PR to yeadon so you can load config files: https://github.com/chrisdembia/yeadon/pull/105
- Move direct collocation system id code into its on repository: https://github.com/csu-hmc/inverted-pendulum-sys-id. I used this method: http://stackoverflow.com/a/359759/467314, as the simpler looking filter branch methods seemed to leave all the history in the new repo instead of just the commits related to the files in the repo.
Last week's issues
- Finish table in the steer torque paper. Update the spreadsheet and insert into the latex doc. -- Did not get to this, but will be priority for this week.
- Add in torque wrench section into the steer torque paper. -- Same as above.
This week's objectives
- [1 hour] Make a blog post explaining how I setup the Plone offsite backups.
- [0.5 hour] Make sure that the hmc.csuohio.edu offsite backup can be restored.
- [3 hours] Finish table in the steer torque paper. Update the spreadsheet and insert into the latex doc.
- [2 hours] Add in torque wrench section into the steer torque paper.
- [0.5 hour] Make a figshare dataset for the torque wrench experiments csv file.
- [4 hours] Send Mont a first draft of the steer torque paper for review (by Saturday).
- [0.5 hour] Try to upload the davis instrumented bicycle h5 file (with raw and processed data) to Figshare.
- [1 hour] Write and introduction to the bicycle data paper.
- [1 hour] Write a section about the raw data in the bicycle data paper.
- [2 hours] Add README's and any other needed data to the davis instrumented bicycle raw data sets and make public.
- [4 hours] Prepare materials for the IEEE open source night (by Friday).
- [0.5 hour] Send out installation instructions for IEEE open source night (by Thursday).
- [2 hours] Pick a talk topic for the seminar and draft an outline of the talk.
- [1 hour] Write system description section in the inverted pendulum paper.
- [5 hours] Implement a ufuncify for vectorized matrix evaluations for the inverted pendulum direct id code. The compile times currently are prohibitive for more than two links.
- [1 hour] Write meta data description in perturbed walking data paper.
- [2 hours] Draft a plan/outline for a new research statement for prof applications.