implementation of the paper "MarginGAN: Adversarial Training in Semi-Supervised Learning"

Last update: Dec 23, 2022

Related tags

Overview

MarginGAN

This repository is the implementation of the paper "MarginGAN: Adversarial Training in Semi-Supervised Learning".

1."preliminary" is the implementation of Preliminary Experiment on MNIST of the paper. Thank the authors of pytorch-generative-model-collections and examples of pytorch, our code is widely adapted from their repositories.

To train the network, an example is as follows:

python main.py \
  --gan_type MarginGAN \
  --num_labels 600 \
  --lrC 0.1 \
  --epoch 50

2."ablation" is the implementation of Ablation Study on MNIST of the paper. To train the network, an example is as follows:

python main.py \
--gan_type MarginGAN_UG \
--num_labels 600 \
--lrC 0.01 \
--epoch 50

3."further" is the implementation of Experiment on SVHN and CIFAR-10 of the paper. Thank the authors of mean teacher, our code is widely adapted from their repositories.

To train the network, an example is as follows:

python MarginGAN_main.py \
    --dataset cifar10 \
    --train-subdir train+val \
    --eval-subdir test \
    --batch-size 128 \
    --labeled-batch-size 31 \
    --arch cifar_shakeshake26 \
    --consistency-type mse \
    --consistency-rampup 5 \
    --consistency 100.0 \
    --logit-distance-cost 0.01 \
    --weight-decay 2e-4 \
    --lr-rampup 0 \
    --lr 0.05 \
    --nesterov True \
    --labels data-local/labels/cifar10/1000_balanced_labels/00.txt  \
    --epochs 180 \
    --lr-rampdown-epochs 210 \
    --ema-decay 0.97 \
    --generated-batch-size 32

implementation of the paper "MarginGAN: Adversarial Training in Semi-Supervised Learning"

Related tags

Overview

MarginGAN

Owner

Van

Scalable machine learning based time series forecasting

A developer interface for creating Chat AIs for the Chai app.

PyTorch implementation of Constrained Policy Optimization

NVTabular is a feature engineering and preprocessing library for tabular data designed to quickly and easily manipulate terabyte scale datasets used to train deep learning based recommender systems.

[NeurIPS 2021] Garment4D: Garment Reconstruction from Point Cloud Sequences

Cosine Annealing With Warmup

PyTorch implementation of the NIPS-17 paper "Poincaré Embeddings for Learning Hierarchical Representations"

Speeding-Up Back-Propagation in DNN: Approximate Outer Product with Memory

A Player for Kanye West's Stem Player. Sort of an emulator.

EMNLP 2021 - Frustratingly Simple Pretraining Alternatives to Masked Language Modeling

Process text, including tokenizing and representing sentences as vectors and Applying some concepts like RNN, LSTM and GRU to create a classifier can detect the language in which a sentence is written from among 17 languages.

Official code base for the poster "On the use of Cortical Magnification and Saccades as Biological Proxies for Data Augmentation" published in NeurIPS 2021 Workshop (SVRHM)

Lite-HRNet: A Lightweight High-Resolution Network

Learning kernels to maximize the power of MMD tests

Codes for Causal Semantic Generative model (CSG), the model proposed in "Learning Causal Semantic Representation for Out-of-Distribution Prediction" (NeurIPS-21)

Official implementation of the paper WAV2CLIP: LEARNING ROBUST AUDIO REPRESENTATIONS FROM CLIP

Code for the paper "JANUS: Parallel Tempered Genetic Algorithm Guided by Deep Neural Networks for Inverse Molecular Design"

MVS2D: Efficient Multi-view Stereo via Attention-Driven 2D Convolutions

Code and results accompanying our paper titled Mixture Proportion Estimation and PU Learning: A Modern Approach at Neurips 2021 (Spotlight)

Source codes for "Structure-Aware Abstractive Conversation Summarization via Discourse and Action Graphs"