Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add matrix-weighted learning #6

Merged
merged 57 commits into from
Jan 19, 2024
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
57 commits
Select commit Hold shift + click to select a range
c43ea3b
Start using sparse solvers
duembgen Dec 28, 2023
4eb7cbd
Merge branch 'master' into use-sparse-solvers
duembgen Dec 29, 2023
3be44fa
Move chompack to pip
duembgen Dec 29, 2023
c003220
Create clique decomposition for stereo
duembgen Dec 30, 2023
c9804c6
Remove redundant sparseqr
duembgen Dec 30, 2023
9f515d1
Fix numerical issues by constraining only first row
duembgen Jan 1, 2024
9904a95
Implement clique form of Wahba and Mono
duembgen Jan 1, 2024
72b545b
Create overlap mode 2
duembgen Jan 2, 2024
975b7a1
Parametrize using all pairs and fusion
duembgen Jan 3, 2024
e443a25
Update submodule
duembgen Jan 3, 2024
5df8095
Fixed for stereo3D
duembgen Jan 3, 2024
6a19785
Merge branch 'master' into use-sparse-solvers
Jan 3, 2024
f11986c
Reproducing results, formatting
Jan 3, 2024
c2429e1
Disable A_known
Jan 3, 2024
55335dc
Fix stereo3D
Jan 3, 2024
0fa2d5f
Merge branch 'use-sparse-solvers' of github.com:utiasASRL/constraint_…
duembgen Jan 3, 2024
6803a5c
Update submodule
Jan 3, 2024
182fd72
Merge branch 'use-sparse-solvers' of github.com:utiasASRL/constraint_…
Jan 3, 2024
c5d6bab
Start improving clique structure
duembgen Jan 4, 2024
c25676c
Fix sparsity issue
Jan 4, 2024
ec72aaa
New clique structure working
Jan 4, 2024
fb1485c
Disentangle overlap and n_vars
Jan 4, 2024
df3b86e
Create customizable template scaling
Jan 5, 2024
7955eb2
Add autotemplate to clique pipeline
Jan 5, 2024
17f315b
Successfully test new pipeline on robust estimators
Jan 6, 2024
5bd5cf8
Clean up template creation
Jan 6, 2024
1f687b5
Implement autotemplate with sparsity
Jan 8, 2024
7c7189c
Implement scaling study
Jan 9, 2024
f9ea78d
Fix stereo2D test and cvxpy Parameters implementation
duembgen Jan 9, 2024
af1c7c7
Cleanup
Jan 9, 2024
bf7b38e
Add mat_weights modules
duembgen Jan 10, 2024
db0a73e
Merge branch 'master' into add_matrix_weighted
duembgen Jan 10, 2024
bf5529c
Sort imports
duembgen Jan 11, 2024
f8e3b10
Fix variable naming
Jan 12, 2024
cbb45c5
first tightness test passing
duembgen Jan 12, 2024
99aa107
Merge branch 'master' into add_matrix_weighted
duembgen Jan 12, 2024
0f43f6a
Update submodule
duembgen Jan 12, 2024
9b4d793
Add matweight lifter
duembgen Jan 12, 2024
42924c2
Merge commit '9b4d793' into add_matrix_weighted
duembgen Jan 12, 2024
16f1031
Update submodule
duembgen Jan 12, 2024
8eb1ae0
Start scalability study
duembgen Jan 12, 2024
27aa056
Merge branch 'add_matrix_weighted' of github.com:utiasASRL/constraint…
duembgen Jan 12, 2024
8600635
Fix import
duembgen Jan 12, 2024
6f31f9a
Merge branch 'master' into use-sparse-solvers
duembgen Jan 12, 2024
ea4d279
Merge branch 'use-sparse-solvers' into add_matrix_weighted
duembgen Jan 12, 2024
1ec1880
Change variable names
Jan 12, 2024
8bf77bb
Add new variables to template creation
Jan 12, 2024
fee3b50
Start clique study
Jan 12, 2024
ec68108
Clique study working
Jan 15, 2024
bac00b8
Generate all plots
Jan 15, 2024
fc9f18a
Fix error
Jan 16, 2024
6517486
Merge branch 'add_matrix_weighted' of github.com:utiasASRL/constraint…
Jan 16, 2024
edcf091
Create timing study
Jan 16, 2024
34a8b52
Integrate ADMMClique
Jan 17, 2024
dd46d77
Add ADMM results
Jan 17, 2024
4213f87
Improve saving
Jan 17, 2024
3926af5
Fix tests
duembgen Jan 19, 2024
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions .github/workflows/python-package-conda.yml
Original file line number Diff line number Diff line change
Expand Up @@ -21,7 +21,7 @@ jobs:
- uses: actions/checkout@v3
with:
token: ${{ secrets.CONSTRAINT_LEARNING_PAT }}
submodules: recursive
submodules: true
- name: Set up Python 3.10
uses: actions/setup-python@v3
with:
Expand All @@ -42,4 +42,4 @@ jobs:
- name: Test with pytest
run: |
conda install pytest
pytest
pytest _test/
2 changes: 2 additions & 0 deletions .gitignore
Original file line number Diff line number Diff line change
@@ -1,5 +1,6 @@
_results/
_results_server/
_results_laptop/
_plots/
starrynight/
log/
Expand All @@ -16,3 +17,4 @@ solve_mosek.ptf
solve_cvxpy*.ptf
build/
dist/
mosek_output.tmp
3 changes: 3 additions & 0 deletions .gitmodules
Original file line number Diff line number Diff line change
Expand Up @@ -4,3 +4,6 @@
[submodule "certifiable-tools"]
path = certifiable-tools
url = [email protected]:utiasASRL/certifiable-tools
[submodule "mat_weight_certs"]
path = mat_weight_certs
url = [email protected]:holmesco/mat_weight_certs
20 changes: 10 additions & 10 deletions _scripts/generate_all_results.py
Original file line number Diff line number Diff line change
@@ -1,21 +1,21 @@
import matplotlib

from _scripts.run_all_study import run_all as run_all_study
from _scripts.run_stereo_study import run_all as run_stereo_study
from _scripts.run_range_only_study import run_all as run_range_only_study
from _scripts.run_other_study import run_all as run_other_study
from _scripts.run_datasets_stereo import run_all as run_datasets_stereo
from _scripts.run_datasets_ro import run_all as run_datasets_ro

import matplotlib
from _scripts.run_datasets_stereo import run_all as run_datasets_stereo
from _scripts.run_other_study import run_all as run_other_study
from _scripts.run_range_only_study import run_all as run_range_only_study
from _scripts.run_stereo_study import run_all as run_stereo_study

try:
# matplotlib.use("TkAgg")
matplotlib.use("Agg")
except:
matplotlib.use("TkAgg")
# matplotlib.use("Agg") # no plotting
except Exception as e:
pass

if __name__ == "__main__":
n_seeds = 1 # was 10
recompute = False
recompute = True
tightness = True
scalability = True

Expand Down
Loading