GitPlanet

Cheap and reliable Node.js hosting starts at $3/month, and $1/month static HTML hosting

Created with love in Canada, visit hostnodejs.com today

Feel like to post an Ad? Learn Details
All Projects → wzell → cmd

wzell / cmd

Licence: other
Central Moment Discrepancy for Domain-Invariant Representation Learning (ICLR 2017, keras)

Programming Languages

python
139335 projects - #7 most used programming language

Labels

neural-networkstransfer-learningdomain-adaptationmoment-alignment

Projects that are alternatives of or similar to cmd

Transferlearning Tutorial
《迁移学习简明手册》LaTex源码
Stars: ✭ 2,122 (+3903.77%)
Mutual labels:  transfer-learning, domain-adaptation
TA3N
[ICCV 2019 Oral] TA3N: https://github.com/cmhungsteve/TA3N (Most updated repo)
Stars: ✭ 45 (-15.09%)
Mutual labels:  transfer-learning, domain-adaptation
Seg Uncertainty
IJCAI2020 & IJCV 2020 🌇 Unsupervised Scene Adaptation with Memory Regularization in vivo
Stars: ✭ 202 (+281.13%)
Mutual labels:  transfer-learning, domain-adaptation
Awesome Transfer Learning
Best transfer learning and domain adaptation resources (papers, tutorials, datasets, etc.)
Stars: ✭ 1,349 (+2445.28%)
Mutual labels:  transfer-learning, domain-adaptation
transfer-learning-algorithms
Implementation of many transfer learning algorithms in Python with Jupyter notebooks
Stars: ✭ 42 (-20.75%)
Mutual labels:  transfer-learning, domain-adaptation
Convolutional Handwriting Gan
ScrabbleGAN: Semi-Supervised Varying Length Handwritten Text Generation (CVPR20)
Stars: ✭ 107 (+101.89%)
Mutual labels:  transfer-learning, domain-adaptation
Clan
( CVPR2019 Oral ) Taking A Closer Look at Domain Shift: Category-level Adversaries for Semantics Consistent Domain Adaptation
Stars: ✭ 248 (+367.92%)
Mutual labels:  transfer-learning, domain-adaptation
Cross Domain ner
Cross-domain NER using cross-domain language modeling, code for ACL 2019 paper
Stars: ✭ 67 (+26.42%)
Mutual labels:  transfer-learning, domain-adaptation
Transfer-learning-materials
resource collection for transfer learning!
Stars: ✭ 213 (+301.89%)
Mutual labels:  transfer-learning, domain-adaptation
pykale
Knowledge-Aware machine LEarning (KALE): accessible machine learning from multiple sources for interdisciplinary research, part of the 🔥PyTorch ecosystem
Stars: ✭ 381 (+618.87%)
Mutual labels:  transfer-learning, domain-adaptation
Ddc Transfer Learning
A simple implementation of Deep Domain Confusion: Maximizing for Domain Invariance
Stars: ✭ 83 (+56.6%)
Mutual labels:  transfer-learning, domain-adaptation
Transformers-Domain-Adaptation
Adapt Transformer-based language models to new text domains
Stars: ✭ 67 (+26.42%)
Mutual labels:  transfer-learning, domain-adaptation
Libtlda
Library of transfer learners and domain-adaptive classifiers.
Stars: ✭ 71 (+33.96%)
Mutual labels:  transfer-learning, domain-adaptation
Shot
code released for our ICML 2020 paper "Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptation"
Stars: ✭ 134 (+152.83%)
Mutual labels:  transfer-learning, domain-adaptation
Deep Transfer Learning
Deep Transfer Learning Papers
Stars: ✭ 68 (+28.3%)
Mutual labels:  transfer-learning, domain-adaptation
Awesome Domain Adaptation
A collection of AWESOME things about domian adaptation
Stars: ✭ 3,357 (+6233.96%)
Mutual labels:  transfer-learning, domain-adaptation
Transfer Learning Library
Transfer-Learning-Library
Stars: ✭ 678 (+1179.25%)
Mutual labels:  transfer-learning, domain-adaptation
Transferlearning
Transfer learning / domain adaptation / domain generalization / multi-task learning etc. Papers, codes, datasets, applications, tutorials.-迁移学习
Stars: ✭ 8,481 (+15901.89%)
Mutual labels:  transfer-learning, domain-adaptation
transfertools
Python toolbox for transfer learning.
Stars: ✭ 22 (-58.49%)
Mutual labels:  transfer-learning, domain-adaptation
BA3US
code for our ECCV 2020 paper "A Balanced and Uncertainty-aware Approach for Partial Domain Adaptation"
Stars: ✭ 31 (-41.51%)
Mutual labels:  transfer-learning, domain-adaptation
View All Similar Projects ➔

CMD

Central Moment Discrepancy for Domain-Invariant Representation Learning - ICLR 2017

Note: An extended journal version is avilable together with its source code and arXiv version

This repository contains code for reproducing the experiments reported in the paper Central Moment Discrepancy (CMD) for Domain-Invariant Representation Learning published at the International Conference on Learning Representations (ICLR2017) by Werner Zellinger, Edwin Lughofer and Susanne Saminger-Platz from the Department of Knowledge Based Mathematical Systems at the JKU Linz, and, Thomas Grubinger and Thomas Natschläger from the Data Analysis Systems Group at the Software Competence Hagenberg.

In the source code, the CMD domain-regularizer is denoted by 'mmatch'.

Requirements

The implementation is based on Theano and the neural networks library Keras. For installing Theano and Keras please follow the installation instruction on the respective github pages. You will also need: numpy, pandas, seaborn, matplotlib, sklearn and scipy

Datasets

We report results for two different benchmark datasets in our paper: AmazonReview and Office. In addition, the model weights of the VGG_16 model pre-trained on Imagenet are used. The AmazonReview data set can be downloaded from http://www.cse.wustl.edu/~mchen/code/mSDA.tar. The file mSDA/examples/amazon.mat should be copied to utils/amazon_dataset/. The Office data set can be downloaded from https://drive.google.com/file/d/0B4IapRTv9pJ1WGZVd1VDMmhwdlE/view. Copy the folders amazon, dslr and webcam to utils/office_dataset/. Download the VGG16 weights file from http://files.heuritech.com/weights/vgg16_weights.h5 and copy it to utils/office_dataset/.

Experiments

Use the files exp_office.py, exp_amazon_review.py and parameter_sensitivity.py to run all the experiments and create all the images from the paper. Please note that the code runs the full grid searches and random restarts and can therefore run some days.

Precomputed weights

If you don't get the exact same results but similar ones, the random number generator of Theano could be the reason. This can be solved by using our precomputed weights in the folder utils/amazon_dataset/precomputed_weights.

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].