issues
search
TorchJD
/
torchjd
Library for Jacobian descent with PyTorch. It enables optimization of neural networks with multiple losses (e.g. multi-task learning).
https://torchjd.org
MIT License
128
stars
0
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Add release.yml action
#146
ValerianRey
closed
21 minutes ago
1
Update version to 0.2.1
#145
ValerianRey
closed
19 hours ago
0
Remove numpy uppercap in pyproject.toml
#144
ValerianRey
closed
19 hours ago
1
Revert IMTLG to using pinv
#143
ValerianRey
closed
1 day ago
1
IMTL-G bug on cuda due to lstsq
#142
ValerianRey
closed
1 day ago
0
Improve IMTLG clarity
#141
ValerianRey
closed
1 day ago
0
Fix IMTLG normalization
#140
ValerianRey
closed
1 day ago
0
Add -i in pdm use command in CONTRIBUTING
#139
ValerianRey
closed
3 days ago
0
Gramian backward exploration
#138
PierreQuinton
closed
21 hours ago
0
High VRAM consumption
#137
lhallee
closed
6 days ago
2
Implement Gramian based aggregators as such
#136
PierreQuinton
opened
6 days ago
1
Use with AMP?
#135
stevendbrown
closed
6 days ago
2
Improve `IMTLG` implementation
#134
ValerianRey
closed
1 week ago
1
Minor fixes
#133
ValerianRey
closed
1 week ago
0
Improve IMTL-G implementation
#132
PierreQuinton
closed
1 week ago
2
Comparison to ConFIG?
#131
ogencoglu
opened
1 week ago
9
Improve README.md
#130
ValerianRey
closed
1 week ago
0
Improve introduction to TorchJD in README.md
#129
ValerianRey
closed
1 week ago
1
Add MTL usage example in README.md
#128
ValerianRey
closed
1 week ago
1
Merge compatibility with installation in README.md
#127
ValerianRey
closed
1 week ago
1
Add aggregator example in README.md
#126
ValerianRey
closed
1 week ago
2
Add table of aggregators in README.md
#125
ValerianRey
closed
1 week ago
1
Improve introduction to the base class Aggregator in doc
#124
ValerianRey
closed
1 week ago
1
Improve introduction to aggregators in the main page of the documentation
#123
PierreQuinton
closed
1 week ago
0
Improve documentation of aggregators
#122
PierreQuinton
closed
1 week ago
1
Introduce the page `aggregation` in `torchjd.org`
#121
PierreQuinton
closed
1 week ago
0
Release candidate 0.2.0
#120
ValerianRey
closed
1 week ago
0
Add parameter `create_graph` to `Grad` and `Jac`
#119
PierreQuinton
closed
1 week ago
0
Revamp `retain_graph` in `Grad` and `Jac`
#118
PierreQuinton
closed
1 week ago
0
Turn `Store` into `Accumulate`
#117
PierreQuinton
closed
1 week ago
0
Turn `Store` into `Accumulate`
#116
ValerianRey
closed
1 week ago
0
Fix useless double assignment in NashMTL
#115
ValerianRey
closed
2 weeks ago
0
Fix types
#114
ValerianRey
closed
2 weeks ago
0
Fix unreliable relative import in interactive_plotter.py
#113
ValerianRey
closed
2 weeks ago
0
Unify indentation in pyproject.toml (4 spaces)
#112
ValerianRey
closed
2 weeks ago
0
Remove unused test dependency contexttimer
#111
ValerianRey
closed
2 weeks ago
0
Improve main doc
#110
ValerianRey
closed
2 weeks ago
0
Add information about compatibility in installation.md
#109
ValerianRey
closed
2 weeks ago
1
Add table to compare aggregators
#108
PierreQuinton
closed
2 weeks ago
1
Improve documentation of parallel_chunk_size
#107
ValerianRey
closed
2 weeks ago
0
Add check that the tensor expects a .grad field in Store._compute.
#106
ValerianRey
closed
2 weeks ago
1
Add link to cpu torch index in pyproject.toml
#105
ValerianRey
closed
3 weeks ago
0
Add cuda tests
#104
ValerianRey
closed
3 weeks ago
0
Test gpu action
#103
ValerianRey
closed
3 weeks ago
1
In the docs, in `aggregation/index.html`, we should add a table of properties.
#102
PierreQuinton
closed
2 weeks ago
2
In `Store` check that the tensor's `.grad` field should be populated
#101
PierreQuinton
closed
2 weeks ago
2
Structure for backwarding gramians
#100
PierreQuinton
opened
1 month ago
2
Cache pdm installation in CI
#99
PierreQuinton
closed
3 weeks ago
4
Remove constraint in `Stack` to have non-empty list of `transform`s
#98
PierreQuinton
closed
2 months ago
1
Run tests on CUDA
#97
ValerianRey
closed
3 weeks ago
0
Next