This repository has been archived by the owner on Sep 27, 2023. It is now read-only.
Feature/assoc demo: New algorithm to perform association analysis using 2 way marginal information and supporting changes. #48
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
positional and named arguments. Add some logging. Add parameter to shell function to remove actual strings from candidates.
- Use Rscript in PATH when running tests
Added TODO.
Current metrics calculated are l1 and l2 norms and a heuristic whether there was a false positive detected (when a spurious candidate string is reported in rappor estimate that doesn't exist in the real distribution).
ProcessAll in analyze.R now returns metrics.
The test parameters are defined in tests/regtest_spec.py. Basic usage is: $ ./regtest.sh run-all This runs all tests in parallel, and results in an HTML table with results. - Calculate both false positives and false negatives in analyze.R. Refactor the function to be more symmetric. - Refactor demo.sh a bit - In gen_sim_input.py, get rid of hard-coded 7 values per client, and make it a parameter - Change rappor_sim.py to use a -d <dist> flag, rather than separate flags - Add test cases based on Chrome params - Factor out util.sh script
- Rename to 'uniform' (probability 1/2) and 'f_mask' (probability f). Rewrite the comments.
Make the Python simulation into a Python client library.
- resolved conflicts - modified code to use new Encode interface - modified rappor_assoc_sim.py to use same interface as rappor_sim.py
Also, some minor refactoring.
- uncommented experimental code in decode2way and documented it - renamed function that processes assoc maps - deleted params.csv
- inverted noise matrix outside loop - renamed gen_assoc_reports - added its test to test.sh - make-summary now shows original dimensions for variables
- threw fitdistribution experimental code into separate function that is now only called by a flag passed to FitDistribution - flag added to assoctest.sh to run comparisons to EM - added package jsonlite to setup - further documentation added in sum_bits_assoc
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Usage: sum_bits_assoc.py
Corresponding test analysis/tools/sum_bits_assoc_test.py tests the python file with detailed documentation of output format
Usage: ./assoctest.sh run-seq ‘^a-’ 5 T
This runs all association tests 5 times comparing both algorithms (T = true flag) in sequential processes. Other options documented in assoctest.sh
Usage: ./quick_assoc.sh [<EM also? T/F>].
Note: directory assumed to have some structure (see documentation in quick_assoc.sh)