remixmatch - GitLite

KO | EN

gitlite — search

#hacktoberfest #ai #typescript #go #machine-learning #rust #deep-learning #linux #docker #macos #nodejs #java #cli #pytorch

remixmatch

(★ 132)

No description available.

파일 탐색기

최종 버전 다운로드 (.zip)

- ab_cta_remixmatch.py
- ab_cta_remixmatch_noweak.py
- ab_remixmatch.py
- - __init__.py
  - train.py
- cta_fsmixup.py
- cta_mixmatch.py
- cta_remixmatch.py
- - __init__.py
  - data.py
  - train.py
- - all.sh
- fs_baseline.py
- fs_mixup.py
- __init__.py
- augment.py
- ctaugment.py
- data.py
- layers.py
- models.py
- train.py
- utils.py
- - ablation.sh
  - cifar10.sh
  - stl10.sh
  - svhn.sh
- aggregate_accuracy.py
- check_split.py
- create_datasets.py
- create_split.py
- create_unlabeled.py
- extract_accuracy.py
- inspect_dataset.py
- - __init__.py
  - augmentations.py
  - custom_ops.py
  - policies.py
  - shake_drop.py
  - shake_shake.py
  - wrn.py
- __init__.py
- LICENSE
- vat_utils.py
CONTRIBUTING.md
ict.py
LICENSE
mean_teacher.py
mixmatch.py
mixup.py
pi_model.py
pseudo_label.py
README.md
remixmatch_no_cta.py
requirements.txt
vat.py

# 설치 가이드

1. 코드 내려받기

git clone https://github.com/google-research/remixmatch

깃허브에서 프로젝트 코드 전체를 내 컴퓨터로 내려받습니다.

cd remixmatch

방금 내려받은 프로젝트 폴더 안으로 이동합니다.

2. 공식 설치 스크립트

쉬움 추천

사전 준비물

APT (Debian/Ubuntu 계열) Debian/Ubuntu 계열 리눅스에 기본 내장된 패키지 매니저입니다.

sudo apt install python3-dev python3-virtualenv python3-tk imagemagick

APT 패키지 저장소에서 바로 설치합니다 (Debian/Ubuntu 계열).

✅ 설치 후 새 터미널을 열고, 프로그램의 버전 확인 명령(예: --version)으로 정상 설치됐는지 확인하세요.

이 레포의 README에 적힌 실제 명령어를 그대로 가져왔습니다.

3. Python

쉬움

사전 준비물

Git GitHub에서 프로젝트 코드를 내려받으려면 필요합니다.
Python 3 설치 시 'Add Python to PATH' 옵션을 꼭 체크하세요 (Windows).

pip install -r requirements.txt

requirements.txt 등에 명시된 파이썬 라이브러리를 설치합니다.

✅ 에러 메시지 없이 실행되고 터미널에 안내 문구가 출력되면 정상입니다.

이 레포의 README에 적힌 실제 명령어를 그대로 가져왔습니다.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# ReMixMatch

Code for the paper: "[ReMixMatch: Semi-Supervised Learning with Distribution Alignment and Augmentation Anchoring](https://arxiv.org/abs/1911.09785)" by David Berthelot, Nicholas Carlini, Ekin D. Cubuk, Alex Kurakin, Kihyuk Sohn, Han Zhang, and Colin Raffel.

This is not an officially supported Google product.

## Setup

**Important**: `ML_DATA` is a shell environment variable that should point to the location where the datasets are installed. See the *Install datasets* section for more details.

### Install dependencies

```bash
sudo apt install python3-dev python3-virtualenv python3-tk imagemagick
virtualenv -p python3 --system-site-packages env3
. env3/bin/activate
pip install -r requirements.txt
```

### Install datasets

```bash
export ML_DATA="path to where you want the datasets saved"
# Download datasets
CUDA_VISIBLE_DEVICES= ./scripts/create_datasets.py
cp $ML_DATA/svhn-test.tfrecord $ML_DATA/svhn_noextra-test.tfrecord

# Create unlabeled datasets
CUDA_VISIBLE_DEVICES= scripts/create_unlabeled.py $ML_DATA/SSL2/svhn $ML_DATA/svhn-train.tfrecord $ML_DATA/svhn-extra.tfrecord &
CUDA_VISIBLE_DEVICES= scripts/create_unlabeled.py $ML_DATA/SSL2/svhn_noextra $ML_DATA/svhn-train.tfrecord &
CUDA_VISIBLE_DEVICES= scripts/create_unlabeled.py $ML_DATA/SSL2/cifar10 $ML_DATA/cifar10-train.tfrecord &
CUDA_VISIBLE_DEVICES= scripts/create_unlabeled.py $ML_DATA/SSL2/cifar100 $ML_DATA/cifar100-train.tfrecord &
CUDA_VISIBLE_DEVICES= scripts/create_unlabeled.py $ML_DATA/SSL2/stl10 $ML_DATA/stl10-train.tfrecord $ML_DATA/stl10-unlabeled.tfrecord &
wait

# Create semi-supervised subsets
for seed in 0 1 2 3 4 5; do
    for size in 40 250 1000 4000; do
        CUDA_VISIBLE_DEVICES= scripts/create_split.py --seed=$seed --size=$size $ML_DATA/SSL2/svhn $ML_DATA/svhn-train.tfrecord $ML_DATA/svhn-extra.tfrecord &
        CUDA_VISIBLE_DEVICES= scripts/create_split.py --seed=$seed --size=$size $ML_DATA/SSL2/svhn_noextra $ML_DATA/svhn-train.tfrecord &
        CUDA_VISIBLE_DEVICES= scripts/create_split.py --seed=$seed --size=$size $ML_DATA/SSL2/cifar10 $ML_DATA/cifar10-train.tfrecord &
    done
    CUDA_VISIBLE_DEVICES= scripts/create_split.py --seed=$seed --size=10000 $ML_DATA/SSL2/cifar100 $ML_DATA/cifar100-train.tfrecord &
    CUDA_VISIBLE_DEVICES= scripts/create_split.py --seed=$seed --size=2500 $ML_DATA/SSL2/cifar100 $ML_DATA/cifar100-train.tfrecord &
    CUDA_VISIBLE_DEVICES= scripts/create_split.py --seed=$seed --size=1000 $ML_DATA/SSL2/stl10 $ML_DATA/stl10-train.tfrecord $ML_DATA/stl10-unlabeled.tfrecord &
    wait
done
CUDA_VISIBLE_DEVICES= scripts/create_split.py --seed=1 --size=5000 $ML_DATA/SSL2/stl10 $ML_DATA/stl10-train.tfrecord $ML_DATA/stl10-unlabeled.tfrecord
```

## Running

### Setup

All commands must be ran from the project root. The following environment variables must be defined:
```bash
export ML_DATA="path to where you want the datasets saved"
export PYTHONPATH=$PYTHONPATH:.
```

### Example

For example, training a remixmatch with 32 filters and 4 augmentations on cifar10 shuffled with `seed=3`, 250 labeled samples and 5000
validation samples:
```bash
CUDA_VISIBLE_DEVICES=0 python cta/cta_remixmatch.py --filters=32 --K=4 --dataset=cifar10.3@250-5000 --w_match=1.5 --beta=0.75 --train_dir ./experiments/remixmatch
```

Available labelled sizes are 40, 100, 250, 1000, 4000.
For validation, available sizes are 1, 5000.
Possible shuffling seeds are 1, 2, 3, 4, 5 and 0 for no shuffling (0 is not used in practiced since data requires to be
shuffled for gradient descent to work properly).

#### Multi-GPU training
Just pass more GPUs and remixmatch automatically scales to them, here we assign GPUs 4-7 to the program:
```bash
CUDA_VISIBLE_DEVICES=4,5,6,7 python cta/cta_remixmatch.py --filters=32 --K=4 --dataset=cifar10.3@250-5000 --w_match=1.5 --beta=0.75 --train_dir ./experiments/remixmatch
```

### Valid dataset names
```bash
for dataset in cifar10 svhn svhn_noextra; do
for seed in 0 1 2 3 4 5; do
for valid in 1 5000; do
for size in 40 250 1000 4000; do
    echo "${dataset}.${seed}@${size}-${valid}"
done; done; done; done

for seed in 0 1 2 3 4 5; do
for valid in 1 5000; do
    echo "cifar100.${seed}@10000-${valid}"
done; done

for seed in 1 2 3 4 5; do
for valid in 1 5000; do
    echo "stl10.${seed}@1000-${valid}"
done; done
echo "stl10.1@5000-1"
```

## Monitoring training progress

You can point tensorboard to the training folder (by default it is `--train_dir=./experiments`) to monitor the training
process:

```bash
tensorboard.sh --port 6007 --logdir experiments
```

## Checkpoint accuracy

We compute the median accuracy of the last 20 checkpoints in the paper, this is done through this code:

```bash
# Following the previous example in which we trained cifar10.3@250-5000, extracting accuracy:
./scripts/extract_accuracy.py experiments/cifar10.d.d.d.3\@250-5000/CTAugment_depth2_th0.80_decay0.990/CTAReMixMatch_K4_archresnet_batch64_beta0.75_filters32_lr0.002_nclass10_redux1st_repeat4_scales3_use_dmTrue_use_xeTrue_w_kl0.5_w_match1.5_w_rot0.5_warmup_kimg1024_wd0.02/
# The command above will create a stats/accuracy.json file in the model folder.
# The format is JSON so you can either see its content as a text file or process it to your liking.
```

## Reproducing tables from the paper

Check the contents of the `runs/*.sh` files, these will give you the commands (and the hyper-parameters) to reproduce the results from the paper.

## Citing this work

```
@article{berthelot2019remixmatch,
    title={ReMixMatch: Semi-Supervised Learning with Distribution Alignment and Augmentation Anchoring},
    author={David Berthelot and Nicholas Carlini and Ekin D. Cubuk and Alex Kurakin and Kihyuk Sohn and Han Zhang and Colin Raffel},
    journal={arXiv preprint arXiv:1911.09785},
    year={2019},
}
```