Adversarially Learned Inference

Last update: Sep 24, 2022

Related tags

Deep Learning ALI

Overview

Adversarially Learned Inference

Code for the Adversarially Learned Inference paper.

Compiling the paper locally

From the repo's root directory,

$ cd papers
$ latexmk --pdf adverarially_learned_inference

Requirements

Blocks, development version
Fuel, development version

Setup

Clone the repository, then install with

$ pip install -e ALI

Downloading and converting the datasets

Set up your ~/.fuelrc file:

$ echo "data_path: \"<MY_DATA_PATH>\"" > ~/.fuelrc

Go to <MY_DATA_PATH>:

$ cd <MY_DATA_PATH>

Download the CIFAR-10 dataset:

$ fuel-download cifar10
$ fuel-convert cifar10
$ fuel-download cifar10 --clear

Download the SVHN format 2 dataset:

$ fuel-download svhn 2
$ fuel-convert svhn 2
$ fuel-download svhn 2 --clear

Download the CelebA dataset:

$ fuel-download celeba 64
$ fuel-convert celeba 64
$ fuel-download celeba 64 --clear

Training the models

Make sure you're in the repo's root directory.

CIFAR-10

$ THEANORC=theanorc python experiments/ali_cifar10.py

SVHN

$ THEANORC=theanorc python experiments/ali_svhn.py

CelebA

$ THEANORC=theanorc python experiments/ali_celeba.py

Toy task

$ THEANORC=theanorc python experiments/ali_mixture.py

$ THEANORC=theanorc python experiments/gan_mixture.py

Evaluating the models

Samples

$ THEANORC=theanorc scripts/sample [main_loop.tar]

e.g.

$ THEANORC=theanorc scripts/sample ali_cifar10.tar

Interpolations

$ THEANORC=theanorc scripts/interpolate [which_dataset] [main_loop.tar]

e.g.

$ THEANORC=theanorc scripts/interpolate celeba ali_celeba.tar

Reconstructions

$ THEANORC=theanorc scripts/reconstruct [which_dataset] [main_loop.tar]

e.g.

$ THEANORC=theanorc scripts/reconstruct cifar10 ali_cifar10.tar

Semi-supervised learning on SVHN

First, preprocess the SVHN dataset with the learned ALI features:

$ THEANORC=theanorc scripts/preprocess_representations [main_loop.tar] [save_path.hdf5]

e.g.

$ THEANORC=theanorc scripts/preprocess_representations ali_svhn.tar ali_svhn_preprocessed.hdf5

Then, launch the semi-supervised script:

$ python experiments/semi_supervised_svhn.py ali_svhn.tar [save_path.hdf5]

e.g.

$ python experiments/semi_supervised_svhn.py ali_svhn_preprocessed.hdf5

[...]
Validation error rate = ... +- ...
Test error rate = ... +- ...

Toy task

$ THEANORC=theanorc scripts/generate_mixture_plots [ali_main_loop.tar] [gan_main_loop.tar]

e.g.

$ THEANORC=theanorc scripts/generate_mixture_plots ali_mixture.tar gan_mixture.tar

Comments

Conditional Generation

I'm interested in getting the update to this codebase that includes the conditional generation, as covered in the more recent version of the paper (related image below). Can you let me know if that will be added to the repo?

opened by dribnet 8
mistake in D(x,z) input size

In table 5 from the paper you state that the input size for D(x,z) is 1024x1x1 which I think it's wrong after looking at the previous output sizes D(x) and D(z). I think that should be 1536x1x1.

Is that assumption correct?

opened by edgarriba 5
deserialization of models hangs

Training goes well for me using the scripts in experiments with the latest version of blocks, but then when I run any subsequent command that uses the generated model like scripts/sample or scripts/reconstruct, the command hangs indefinitely. My guess is that the deserialization is getting jammed up.

I can look into it more - not yet familiar with the new tar format - but curious if this might be a known issue.

opened by dribnet 3
Fuel version problem

I installed the current development version of fuel, but had some issue in fuel downloading.

$ fuel-download celeba 64 $ fuel-convert celeba 64 $ fuel-download celeba 64 --clear

The error message I got is: fuel-download: error: unrecognized arguments: 64 if I remove 64, I got: TypeError: init() got an unexpected keyword argument 'max_value'

Could someone please specify what version or commits of fuel and progressbar should I use? Thanks

opened by hope-yao 1
Where to use the reparametrization trick

In the decoder module. I found that z is sampled from N(0, 1), so where did you use the reparametrization trick described in formual (2) and (3) in the paper

opened by wuhaozhe 0
semi-supervised learning

Hello,I read the paper and the source code.And it mentioned 'The last three hidden layers of the encoder as well as its output are concatenated to form a 8960-dimensional feature vector.' in section 4.3 of the paper.Could you please tell me how to compute the dimension?Thanks very much

opened by C-xiaomeng 1

ImportError: No module named ali.utils

I followed the same steps in the readme file, but when I run this line

$ THEANORC=theanorc python experiments/ali_cifar10.py

I get:

Traceback (most recent call last):
  File "experiments/ali_cifar10.py", line 3, in <module>
    from ali.utils import get_log_odds, conv_brick, conv_transpose_brick, bn_brick
ImportError: No module named ali.utils

opened by xtarx 0

Preprocess_representation has a bug for me

Hi, I was trying to reproduce the representation learning results of paper. Everything works fine except "preprocess_representations" script. It is leading to this error:

File "scripts/preprocess_representations", line 32, in preprocess_svhn bricks=[ali.encoder.layers[-9], ali.encoder.layers[-6], AttributeError: 'GaussianConditional' object has no attribute 'layers'

Any help would be appreciated.

opened by MarziEd 1
Semi-supervised learning

I've been trying to reproduce your figures for semi-supervised learning on CIFAR-10 (19.98% with 1000 labels). This result is based on the technique proposed in Salimans et al. (2016), not SVMs. Is there any way you can include your code, or at least any changes to the hyperparameters in ali_cifar10.py?

Thanks in advance for your help.

opened by christiancosgrove 7

Releases(v1)

v1(Jun 3, 2016)

Source code(tar.gz)
Source code(zip)

Owner

Mohamed Ishmael Belghazi

GitHub Repository

This is an official implementation of our CVPR 2021 paper "Bottom-Up Human Pose Estimation Via Disentangled Keypoint Regression" (https://arxiv.org/abs/2104.02300)

Bottom-Up Human Pose Estimation Via Disentangled Keypoint Regression Introduction In this paper, we are interested in the bottom-up paradigm of estima

367 Dec 27, 2022

95.47% on CIFAR10 with PyTorch

Train CIFAR10 with PyTorch I'm playing with PyTorch on the CIFAR10 dataset. Prerequisites Python 3.6+ PyTorch 1.0+ Training # Start training with: py

5k Dec 30, 2022

The toolkit to generate auto labeled datasets

Ozeu Ozeu is the toolkit to autolabal dataset for instance segmentation. You can generate datasets labaled with segmentation mask and bounding box fro

28 Mar 28, 2022

Adversarial-Information-Bottleneck - Distilling Robust and Non-Robust Features in Adversarial Examples by Information Bottleneck (NeurIPS21)

NeurIPS 2021 Title: Distilling Robust and Non-Robust Features in Adversarial Exa

35 Dec 26, 2022

Plugin adapted from Ultralytics to bring YOLOv5 into Napari

napari-yolov5 Plugin adapted from Ultralytics to bring YOLOv5 into Napari. Training and detection can be done using the GUI. Training dataset must be

2 May 05, 2022

AirPose: Multi-View Fusion Network for Aerial 3D Human Pose and Shape Estimation

AirPose AirPose: Multi-View Fusion Network for Aerial 3D Human Pose and Shape Estimation Check the teaser video This repository contains the code of A

41 Dec 05, 2022

This code uses generative adversarial networks to generate diverse task allocation plans for Multi-agent teams.

Mutli-agent task allocation This code uses generative adversarial networks to generate diverse task allocation plans for Multi-agent teams. To change

5 Oct 12, 2022

Deep Learning Interviews book: Hundreds of fully solved job interview questions from a wide range of key topics in AI.

This book was written for you: an aspiring data scientist with a quantitative background, facing down the gauntlet of the interview process in an increasingly competitive field. For most of you, the

4.1k Dec 28, 2022

Official Pytorch implementation of "DivCo: Diverse Conditional Image Synthesis via Contrastive Generative Adversarial Network" (CVPR'21)

DivCo: Diverse Conditional Image Synthesis via Contrastive Generative Adversarial Network Pytorch implementation for our DivCo. We propose a simple ye

64 Nov 22, 2022

Vis2Mesh: Efficient Mesh Reconstruction from Unstructured Point Clouds of Large Scenes with Learned Virtual View Visibility ICCV2021

Vis2Mesh This is the offical repository of the paper: Vis2Mesh: Efficient Mesh Reconstruction from Unstructured Point Clouds of Large Scenes with Lear

71 Dec 25, 2022

Adversarially Learned Inference

Related tags

Overview

Adversarially Learned Inference

Compiling the paper locally

Requirements

Setup

Downloading and converting the datasets

Training the models

CIFAR-10

SVHN

CelebA

Toy task

Evaluating the models

Samples

Interpolations

Reconstructions

Semi-supervised learning on SVHN

Toy task

Comments

Releases(v1)

v1(Jun 3, 2016)

Owner

Mohamed Ishmael Belghazi

This is an official implementation of our CVPR 2021 paper "Bottom-Up Human Pose Estimation Via Disentangled Keypoint Regression" (https://arxiv.org/abs/2104.02300)

95.47% on CIFAR10 with PyTorch

The toolkit to generate auto labeled datasets

Adversarial-Information-Bottleneck - Distilling Robust and Non-Robust Features in Adversarial Examples by Information Bottleneck (NeurIPS21)

Plugin adapted from Ultralytics to bring YOLOv5 into Napari

AirPose: Multi-View Fusion Network for Aerial 3D Human Pose and Shape Estimation

This code uses generative adversarial networks to generate diverse task allocation plans for Multi-agent teams.

Deep Learning Interviews book: Hundreds of fully solved job interview questions from a wide range of key topics in AI.

Official Pytorch implementation of "DivCo: Diverse Conditional Image Synthesis via Contrastive Generative Adversarial Network" (CVPR'21)

🇰🇷 Text to Image in Korean

Model of an AI powered sign language interpreter.

unet-family: Ultimate version

Physics-Aware Training (PAT) is a method to train real physical systems with backpropagation.

Computer-Vision-Paper-Reviews - Computer Vision Paper Reviews with Key Summary along Papers & Codes

Align before Fuse: Vision and Language Representation Learning with Momentum Distillation

basic tutorial on pytorch

RSNA Intracranial Hemorrhage Detection with python

Ranger deep learning optimizer rewrite to use newest components

PaddleBoBo是基于PaddlePaddle和PaddleSpeech、PaddleGAN等开发套件的虚拟主播快速生成项目

Vis2Mesh: Efficient Mesh Reconstruction from Unstructured Point Clouds of Large Scenes with Learned Virtual View Visibility ICCV2021