-
Notifications
You must be signed in to change notification settings - Fork 1
/
gameplan.txt
71 lines (63 loc) · 3.5 KB
/
gameplan.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
# --- Goals ---
- Revised "swdft" package for thesis on CRAN by defense
# --- Tasks ---
## General
- Polish the README and update so that it references thesis
## R Package
### TODO: Version 1.0.0
- Submit to CRAN!
+ Get R CMD CHECK passing! (DONE)
+ Get travis.yaml working (DONE)
+ Re-write: DESCRIPTION file w/ new version (DONE)
+ Update: the README.md file in the base repository to remove the SWDFT for time-series paper (DONE)
+ Write: Vignette that displays the basic functionality
* Add all code to the document (DONE)
* Make sure devtools::check() still passes (DONE)
* Fill in the document w/ a first draft of writing.
- Add writing to SWDFT + Viz (DONE)
- Add Writing to Cosine Regression (DONE)
- Add writing to Demodulation (DONE)
+ Say something about threshold in matching demodulation (DONE)
* Add coefficients(), fitted(), and residuals() calls to cosine regression. (DONE)
* Add formula's to the swdft, cosine regression, and local cosine (DONE) regression to make the exposition more clear. (DONE)
* Edit, Edit, Edit! (DONE)
* Re-send a draft to Sam (DONE)
* Add a brief description of the plot (DONE)
- In the SWDFT plot. Explain the x and y axis, and how they correspond to the mathematical definition. (DONE)
* Add a reference to Kass' book on the tapering / smoothing section. (DONE)
- Add a reference to everything on that section, and link to the PDF copy of the book online (DONE)
* add swdft::get_max_dft(), get_max_swdft() functions (DONE)
* Read through one more time (DONE)
* Mention what I like about the FFT: raw coefficients (DONE)
* Send to committee + Francesca / Taylor / Yotam / etc (DONE)
* Add additional paragraph intro. (DONE)
- Time (ARMA) / Frequency domain (DFT, Oscillations) (DONE)
* I use signals / time-series interchangeably (DONE)
* Should I move the SWDFT introduction to that subsection? Maybe put the SWDFT intro before introducing the local cosine signal? (DONE)
* Figure out how to add references (DONE)
* Read through Francesca's edits. (DONE)
* Make sure the win-builder WARNING is fixed! (DONE)
* (MAYBE) label the red / blue dots (DONE)
* (MAYBE) mention something about padding the SWDFT! (DONE)
* Read one more time (DONE)
* Fix: .Rmd issue on front of the Github page. (DONE)
*
+ Update the version number to 1.0.1 when finished! And also TAG the Github repository.
### TODO: Version 1.1.0
- Change the default visualization to "fraction"
- Handle the case where a$a is passed.
- add examples to documentation for the main functions.
- Cosine regression: Add a "number of parameters" argument.
- Cosine regression: Add an option that automatically selects the largest frequency, use nloptR to find it specifically!
- Complex demod + matching demod: Add: Option to pass in a custom range of frequencies. To this AND to complex_demod!
+ Also, a custom range of pass-frequencies / moving average orders!
- See if I can write the entire algorithm in c++, then only use a wrapper function for the entire method. This would allow easy integration with Python and other languages as well.
- Add a customized Multitaper method.
## Python Package
- Read, understand Python Packages and Modules
+ Create leePy for examples of how this works!
- Read, understand "PyTest" Testing Framework
- Read, understand Sphinx Documentation
- Read, understand virtual environments
## **C++ Package**
- Figure out how to put the "swdft" function entirely in C++, only wrappers.