Multi-Glimpse Network With Python

Last update: May 10, 2022

Related tags

Deep Learning MGNet

Overview

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.

Multi-Glimpse Network With Python

Related tags

Overview

Multi-Glimpse Network

Installation

Evaluation

Accuracy on clean images

Accuracy on adversarial attacks (PGD)

Accuracy on common corruptions

Training

Owner

Local Attention - Flax module for Jax

Emotion classification of online comments based on RNN

Domain Adaptation with Invariant RepresentationLearning: What Transformations to Learn?

The Few-Shot Bot: Prompt-Based Learning for Dialogue Systems

A high-performance distributed deep learning system targeting large-scale and automated distributed training.

Image segmentation with private İstanbul Dataset

LieTransformer: Equivariant Self-Attention for Lie Groups

Pytorch implementation of One-Shot Affordance Detection

Change is Everywhere: Single-Temporal Supervised Object Change Detection in Remote Sensing Imagery (ICCV 2021)

Train a state-of-the-art yolov3 object detector from scratch!

Transfer SemanticKITTI labeles into other dataset/sensor formats.

Python inverse kinematics for your robot model based on Pinocchio.

SAS output to EXCEL converter for Cornell/MIT Language and acquisition lab

CLIP (Contrastive Language–Image Pre-training) trained on Indonesian data

Continuous Time LiDAR odometry

Self-training for Few-shot Transfer Across Extreme Task Differences

Power Core Simulator!

Gradient Inversion with Generative Image Prior

Implementation of "Fast and Flexible Temporal Point Processes with Triangular Maps" (Oral @ NeurIPS 2020)

Combine Tacotron2 and Hifi GAN to generate speech from text