taalbrecht / discretechoicedesigner Goto Github PK
View Code? Open in Web Editor NEWD-optimal design creation tool for discrete choice experiments
D-optimal design creation tool for discrete choice experiments
Add feature to "find" best design with regards to number of questions & number of alternates. After running, plot d-efficiency of each model vs. the number of runs and vs the number of alternates. Ask user to estimate maximum number of alternates, questions, and questions per respondent.
Allow users to enter comma separated values to name all levels for attribute parameters.
Can calculate as lowest common denominator of all possible variable levels in model matrix
Unrecoverable error that can be replicated by using the following settings
2 continuous inputs with 5 levels from -1 to 1 with A+B+A*B as the input formula for each one. Error does not occur if only 2 formulas are used.
Trend of error occurring during the first model run for each session, both in the single and multi design tabs. Cause is unknown. Subsequent runs beyond the first attempt seem to be successful and do not have the same issue. Specified design parameters were identical between the first attempt and subsequent attempts in the observed occurrences of this bug.
Tried to make an experiment with three variables, A, B, and C with B as attribute and A and C as continuous with 2 levels and min/max of 0,1 and -1, 1 respectively. Two formulas were used:
A+B+C - weight 0.4
A+B+I(A^2) - weight 0.6
All parameter priors set to 0
2 alternates
6 questions
1 randomized start
If A has 2 levels, this will cause an error as A^2 and A must be perfectly correlated in the second function. C can be set to two levels with no problem.
Add error statement when this occurs instead of causing a random untraceable error from the users perspective
If the effect size and difference to detect fields are populated, they clear their values when switching between tabs in some cases. The other dynamically sized input fields on the same page are not affected (min and max, number of levels, etc)
Add ability for user to select which type of optimization to use for each formula. Jointly optimize between weighting parameter and optimization type for each equation.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.