motiwari BanditPAM issues

motiwari / BanditPAM

BanditPAM C++ implementation and Python package

MIT License

647 stars 38 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

#include carma only for Python

#221 motiwari opened 1 year ago
0
Possible clashing of compiler opts

#220 motiwari opened 1 year ago
0
Fix compiler flags

#219 motiwari opened 1 year ago
0
When compiling mac wheels, add `'-framework Accelerate'` to compiler flags

#218 motiwari opened 1 year ago
0
Address sanitizer detects memory leaks with `new` for `cache`

#217 motiwari opened 1 year ago
0
Make an enum for sample complexity category

#216 motiwari opened 1 year ago
0
Clean up scripts in `scripts`

#215 motiwari opened 1 year ago
0
Implement `getTotalSwapTime` and `getTimePerSwap`

#214 motiwari opened 1 year ago
0
Make sure wheels aren't being built with `-g` flag

#213 motiwari opened 1 year ago
0
Some compiler flags are duplicated

#212 motiwari opened 1 year ago
0
Bug when `max_iter=1` is passed

#211 motiwari opened 1 year ago
0
Refactor code to make C++ code in top-level and each language binding in subdirectories

#210 motiwari opened 1 year ago
0
Transitioning to a single code base for R/Python

#209 bnaras opened 1 year ago
0
Implement a `make clean`

#208 motiwari opened 1 year ago
0
User report: docker installation instructions in `README` are incomplete

#207 motiwari opened 1 year ago
0
Fix platform-specific guide links in `README`

#206 motiwari opened 1 year ago
0
Easily compare the effects of cache and permutation with flags

#205 lukeleeai opened 1 year ago
1
Idea: from first batch of ref points in each swap step, only keep track of best medoid to flip with (probably the one it's assigned to?)

#204 motiwari opened 1 year ago
0
Both `main` and `slowdown` branches appear to be nondeterministic unless number of threads is set to 1

#203 motiwari opened 1 year ago
0
Stop SWAP steps after they improve loss by less than some threshold $t$ (e.g., 0.1%)

#202 motiwari opened 1 year ago
0
Allow for $\epsilon$-suboptimal arms

#201 motiwari opened 1 year ago
0
Need to ensure `k` elements broadcasting to for each arm are laid out contiguously in memory (after `slowdown` merge)

#200 motiwari opened 1 year ago
0
minor fix

#199 motiwari closed 1 year ago
1
`useCacheP=True` & `usePerm=False` runs slower than BanditPAM with no caching when the dataset is large

#198 lukeleeai opened 1 year ago
0
Upgrade `carma`

#197 motiwari opened 1 year ago
0
Update `README.md`

#196 motiwari opened 1 year ago
0
Might save a factor of 2: in first batch, for arms in which CIs are wide compared to delta mean, predict that we'll need to compute exactly and do so

#195 motiwari opened 1 year ago
0
Fix link to examples on `README`

#194 motiwari opened 2 years ago
0
Need to compile wheels for other architectures for manylinux

#193 motiwari opened 2 years ago
0
Testing slack notifications 2

#192 motiwari closed 2 years ago
0
Testing slack notifications

#191 motiwari closed 2 years ago
0
Categorical features in banditPAM

#190 retzerjj closed 1 year ago
1
Undefined symbol

#189 SergioG-M opened 2 years ago
0
Missing data files

#188 rheinonen closed 1 year ago
1
Attempting to resolve build issues on some M1 Macs

#187 motiwari opened 2 years ago
0
Fix readthedocs link on `README.md`

#186 motiwari opened 2 years ago
0
predict function for Python implementation?

#185 m-muaz opened 2 years ago
3
BanditPAM `v3.0.4`

#184 motiwari closed 2 years ago
0
BanditPAM v4.0.0

#183 motiwari closed 1 year ago
0
cosine similarity vs cosine distance

#182 tazitoo closed 2 years ago
3
BanditPAM `v3.0.3`

#181 motiwari closed 2 years ago
0
BanditPAM `v3.0.3`

#180 motiwari closed 2 years ago
1
Much Slower than Scikit-Learn kMedoids on Large Dimensions

#179 kayuksel closed 2 years ago
2
BanditPAM `v3.0.2`

#178 motiwari closed 2 years ago
0
Update "full suite of tests" command

#177 motiwari closed 2 years ago
1
Allow choosing the random seed (from python)

#176 kno10 closed 2 years ago
2
BanditPAM 200x slower than quadratic algorithms at 10k MNIST

#175 kno10 opened 2 years ago
5
Return loss (cost) of the solution found

#174 kno10 closed 2 years ago
1
BanditPAM crashes colab kernel when invalid loss function name is used

#173 kno10 closed 2 years ago
1
Remove some unnecessary dependencies

#172 motiwari opened 2 years ago
0

Previous Next