Code for the paper "Regularizing Variational Autoencoder with Diversity and Uncertainty Awareness"

Related tags

Deep LearningDU-VAE
Overview

DU-VAE

This is the pytorch implementation of the paper "Regularizing Variational Autoencoder with Diversity and Uncertainty Awareness"

Acknowledgements

Our code is mainly based on this public code. Very thanks for its authors.

Requirements

  • Python >= 3.6
  • Pytorch >= 1.5.0

Data

Datastes used in this paper can be downloaded in this link, with the specific license if that is not based on MIT License.

Usage

Example script to train DU-VAE on text data:

python text.py --dataset yelp \
 --device cuda:0  \
--gamma 0.5 \
--p_drop 0.2 \
--delta_rate 1 \
--kl_start 0 \
--warm_up 10

Example script to train DU-VAE on image data:

python3.6 image.py --dataset omniglot \
 --device cuda:3 \
--kl_start 0 \
--warm_up 10 \
--gamma 0.5  \
--p_drop 0.1 \
--delta_rate 1 \
--dataset omniglot

Example script to train DU-IAF, a variant of DU-VAE, on text data:

python3.6 text_IAF.py --device cuda:2 \
--dataset yelp \
--gamma 0.6 \
--p_drop 0.3 \
--delta_rate 1 \
--kl_start 0 \
--warm_up 10 \
--flow_depth 2 \
--flow_width 60

Example script to train DU-IAF on image data:

python3.6 image_IAF.py --dataset omniglot\
  --device cuda:3 \
--kl_start 0 \
--warm_up 10 \
--gamma 0.5 \
 --p_drop 0.15\
 --delta_rate 1 \
--flow_depth 2\
--flow_width 60 

Here,

  • --dataset specifies the dataset name, currently it supports synthetic, yahoo, yelp for text.py and omniglot for image.py.
  • --kl_start represents starting KL weight (set to 1.0 to disable KL annealing)
  • --warm_up represents number of annealing epochs (KL weight increases from kl_start to 1.0 linearly in the first warm_up epochs)
  • --gamma represents the parameter $\gamma$ in our Batch-Normalization approach, which should be more than 0 to use our model.
  • --p_drop represents the parameter $1-p$ in our Dropout approach, which denotes the percent of data to be ignored and should be ranged in (0,1).
  • --delta_rate represents the hyper-parameter $\alpha$ to controls the min value of the variance $\delta^2$
  • --flow_depth represents number of MADE layers used to implement DU-IAF.
  • --flow_wdith controls the hideen size in each IAF block, where we set the product between the value and the dimension of $z$ as the hidden size. For example, when we set --flow width 60 with the dimension of $z$ as 32, the hidden size of each IAF block is 1920.

Reference

If you find our methods or code helpful, please kindly cite the paper:

@inproceedings{shen2021regularizing,
  title={Regularizing Variational Autoencoder with Diversity and Uncertainty Awareness},
  author={Shen, Dazhong  and Qin, Chuan and Wang, Chao and Zhu, Hengshu and Chen, Enhong and Xiong, Hui},
  booktitle={Proceedings of the 30th International Joint Conference on Artificial Intelligence (IJCAI-21)},
  year={2021}
}
Owner
Dazhong Shen
Dazhong Shen
Sequence to Sequence (seq2seq) Recurrent Neural Network (RNN) for Time Series Forecasting

Sequence to Sequence (seq2seq) Recurrent Neural Network (RNN) for Time Series Forecasting Note: You can find here the accompanying seq2seq RNN forecas

Guillaume Chevalier 1k Dec 25, 2022
FasterAI: A library to make smaller and faster models with FastAI.

Fasterai fasterai is a library created to make neural network smaller and faster. It essentially relies on common compression techniques for networks

Nathan Hubens 193 Jan 01, 2023
Code for paper "Learning to Reweight Examples for Robust Deep Learning"

learning-to-reweight-examples Code for paper Learning to Reweight Examples for Robust Deep Learning. [arxiv] Environment We tested the code on tensorf

Uber Research 261 Jan 01, 2023
Tensorflow implementation of soft-attention mechanism for video caption generation.

SA-tensorflow Tensorflow implementation of soft-attention mechanism for video caption generation. An example of soft-attention mechanism. The attentio

Paul Chen 153 Nov 14, 2022
Have you ever wondered how cool it would be to have your own A.I

Have you ever wondered how cool it would be to have your own A.I. assistant Imagine how easier it would be to send emails without typing a single word, doing Wikipedia searches without opening web br

Harsh Gupta 1 Nov 09, 2021
Code of the paper "Part Detector Discovery in Deep Convolutional Neural Networks" by Marcel Simon, Erik Rodner and Joachim Denzler

Part Detector Discovery This is the code used in our paper "Part Detector Discovery in Deep Convolutional Neural Networks" by Marcel Simon, Erik Rodne

Computer Vision Group Jena 17 Feb 22, 2022
Codes for CVPR2021 paper "PWCLO-Net: Deep LiDAR Odometry in 3D Point Clouds Using Hierarchical Embedding Mask Optimization"

PWCLO-Net: Deep LiDAR Odometry in 3D Point Clouds Using Hierarchical Embedding Mask Optimization (CVPR 2021) This is the official implementation of PW

Intelligent Robotics and Machine Vision Lab 42 Dec 18, 2022
Colar: Effective and Efficient Online Action Detection by Consulting Exemplars, CVPR 2022.

Colar: Effective and Efficient Online Action Detection by Consulting Exemplars This repository is the official implementation of Colar. In this work,

LeYang 246 Dec 13, 2022
Code for the paper "A Study of Face Obfuscation in ImageNet"

A Study of Face Obfuscation in ImageNet Code for the paper: A Study of Face Obfuscation in ImageNet Kaiyu Yang, Jacqueline Yau, Li Fei-Fei, Jia Deng,

35 Oct 04, 2022
Serve TensorFlow ML models with TF-Serving and then create a Streamlit UI to use them

TensorFlow Serving + Streamlit! ✨ 🖼️ Serve TensorFlow ML models with TF-Serving and then create a Streamlit UI to use them! This is a pretty simple S

Álvaro Bartolomé 18 Jan 07, 2023
A simple program for training and testing vit

Vit This is a simple program for training and testing vit. Key requirements: torch, torchvision and timm. Dataset I put 5 categories of the cub classi

xiezhenyu 2 Oct 11, 2022
A pure PyTorch implementation of the loss described in "Online Segment to Segment Neural Transduction"

ssnt-loss ℹ️ This is a WIP project. the implementation is still being tested. A pure PyTorch implementation of the loss described in "Online Segment t

張致強 1 Feb 09, 2022
Picasso: A CUDA-based Library for Deep Learning over 3D Meshes

The Picasso Library is intended for complex real-world applications with large-scale surfaces, while it also performs impressively on the small-scale applications over synthetic shape manifolds. We h

97 Dec 01, 2022
Ludwig is a toolbox that allows to train and evaluate deep learning models without the need to write code.

Translated in 🇰🇷 Korean/ Ludwig is a toolbox that allows users to train and test deep learning models without the need to write code. It is built on

Ludwig 8.7k Jan 05, 2023
PyTorch implementation of "Simple and Deep Graph Convolutional Networks"

Simple and Deep Graph Convolutional Networks This repository contains a PyTorch implementation of "Simple and Deep Graph Convolutional Networks".(http

chenm 253 Dec 08, 2022
Fast image augmentation library and an easy-to-use wrapper around other libraries

Albumentations Albumentations is a Python library for image augmentation. Image augmentation is used in deep learning and computer vision tasks to inc

11.4k Jan 09, 2023
Use graph-based analysis to re-classify stocks and to improve Markowitz portfolio optimization

Dynamic Stock Industrial Classification Use graph-based analysis to re-classify stocks and experiment different re-classification methodologies to imp

Sheng Yang 10 Dec 05, 2022
Human Detection - Pedestrian Detection using OpenCV Python

Pedestrian Detection using OpenCV Python Follow us on Instagram for Machine Lear

Hrishikesh Dutta 1 Jan 23, 2022
Pytorch implementation of "Training a 85.4% Top-1 Accuracy Vision Transformer with 56M Parameters on ImageNet"

Token Labeling: Training an 85.4% Top-1 Accuracy Vision Transformer with 56M Parameters on ImageNet (arxiv) This is a Pytorch implementation of our te

蒋子航 383 Dec 27, 2022
Huawei Hackathon 2021 - Sweden (Stockholm)

huawei-hackathon-2021 Contributors DrakeAxelrod Challenge Requirements: python=3.8.10 Standard libraries (no importing) Important factors: Data depend

Drake Axelrod 32 Nov 08, 2022