Multi-Glimpse Network
Our code requires Python ≥ 3.8
Installation
For example, venv + pip:
$ python3 -m venv env
$ source env/bin/activate
(env) $ python3 -m pip install -r requirements.txt
Evaluation
Accuracy on clean images
- Create ImageNet100 from ImageNet (using symbolic links).
$ python3 tools/create_imagenet100.py tools/imagenet100.txt \
/path/to/ImageNet /path/to/ImageNet100
-
Download checkpoints from Google Drive.
-
Test accuracy.
$ export dataset="--train_dir /path/to/ImageNet100/train \
--val_dir /path/to/ImageNet100/val \
--dataset imagenet --num_class 100"
# Baseline
$ python3 main.py $dataset --test --n_iter 1 --scale 1.0 --model resnet18 \
--checkpoint resnet18_baseline
# Ours
$ python3 main.py $dataset --test --n_iter 4 --scale 2.33 --model resnet18 \
--checkpoint resnet18_ours --alpha 0.6 --s 0.02
Add the flag --flop_count to count the approximate FLOPs for the inference of an image. (using fvcore)
Accuracy on adversarial attacks (PGD)
- Test adversarial accuracy.
# Baseline
$ python3 main.py $dataset --test --n_iter 1 --scale 1.0 --adv --step_k 10 \
--model resnet18 --checkpoint resnet18_baseline
# Ours
$ python3 main.py $dataset --test --n_iter 4 --scale 2.33 --adv --step_k 10 \
--model resnet18 --checkpoint resnet18_ours --alpha 0.6 --s 0.02
Accuracy on common corruptions
- Create ImageNet100-C from ImageNet-C (using symbolic links).
$ python3 tools/create_imagenet100c.py \
tools/imagenet100.txt /path/to/ImageNet-C/ /path/to/ImageNet100-C/
- Test for a single corruption.
$ export dataset="--train_dir /path/to/ImageNet100/train \
--val_dir /path/to/ImageNet100-C/pixelate/5 \
--dataset imagenet --num_class 100"
# Baseline
$ python3 main.py $dataset --test --n_iter 1 --scale 1.0 --model resnet18 \
--checkpoint resnet18_baseline
# Ours
$ python3 main.py $dataset --test --n_iter 4 --scale 2.33 --model resnet18 \
--checkpoint resnet18_ours --alpha 0.6 --s 0.02
- A simple script to test all corruptions and collect results.
# Modify tools/eval_imagenet100c.py and run it to generate script
$ python3 tools/eval_imagenet100c.py /home2/ImageNet100-C/ > run.sh
# Evaluate
$ bash run.sh
# Collect results
$ python3 tools/collect_imagenet100c.py
Training
$ export dataset="--train_dir /path/to/ImageNet100/train \
--val_dir /path/to/ImageNet100/val \
--dataset imagenet --num_class 100"
# Baseline
$ python3 main.py $dataset --epochs 400 --n_iter 1 --scale 1.0 \
--model resnet18 --gpu 0,1,2,3
# Ours
$ python3 main.py $dataset --epochs 400 --n_iter 4 --scale 2.33 \
--model resnet18 --alpha 0.6 --s 0.02 --gpu 0,1,2,3
Check tensorboard for the logs. (When training with multiple gpus, the log value may be scaled by the number of gpus except for the validation accuracy)
tensorboard --logdir=logs
Note that we left our exploration in the code for further study, e.g., self-supervised spatial guidance, dynamic gradient re-scaling operation.
28 Nov 06, 2022
126 Dec 05, 2022
3 Mar 31, 2022
11 Sep 10, 2022
21 Jan 02, 2022
7 Aug 08, 2022
6.7k Dec 28, 2022
3 Dec 05, 2022
109 Dec 29, 2022
0 Mar 27, 2022
43 Dec 16, 2022
87 Dec 16, 2022
33 Nov 24, 2022
22 Mar 16, 2022
6 Dec 05, 2022
2 Apr 28, 2022
7 Jan 24, 2022
65 Dec 20, 2022
4.9k Dec 31, 2022
399 Dec 28, 2022