issues
search
MadryLab
/
trak
A fast, effective data attribution method for neural networks in PyTorch
https://trak.csail.mit.edu/
MIT License
169
stars
22
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Fix a bit typo README
#18
guspan-tanadi
closed
1 year ago
0
0.1.1
#17
kristian-georgiev
closed
1 year ago
0
Tutorial for CLIP model output
#16
kristian-georgiev
closed
1 year ago
1
NLP model output
#15
kristian-georgiev
closed
1 year ago
1
Concurrent TRAKer init (with diff model id) in same save dir may lead to I/O errors
#14
kristian-georgiev
closed
1 year ago
1
Do not allocate train x target score matrix for each model ID
#13
kristian-georgiev
closed
1 year ago
1
Make `fast_jl` a separate package, add it as an optional dependency to `trak`
#11
kristian-georgiev
closed
1 year ago
1
Implement soft thresholding
#10
kristian-georgiev
closed
1 year ago
2
Add `start_scoring` to simplify keeping track of target gradients
#9
kristian-georgiev
closed
1 year ago
2
Fix model ID loading
#8
bencw99
closed
1 year ago
0
Redesign API
#7
kristian-georgiev
closed
1 year ago
1
merge fast JL code
#6
kristian-georgiev
closed
1 year ago
1
sklearn-style API for image classification models
#5
kristian-georgiev
closed
1 year ago
0
Make `block_size` adapt to available RAM/GPU VRAM
#4
kristian-georgiev
closed
1 year ago
1
Proper docstrings
#3
kristian-georgiev
closed
1 year ago
6
Implement `zarr` saver
#2
kristian-georgiev
closed
1 year ago
0
Allow taking gradients with respect to a subset of the model's parameters
#1
kristian-georgiev
closed
10 months ago
4
Previous