Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
zhaijianyang cae450be12 | 2 years ago | |
---|---|---|
data | 2 years ago | |
.gitignore | 2 years ago | |
DEC.py | 2 years ago | |
IDEC.py | 2 years ago | |
README.md | 2 years ago | |
datasets.py | 2 years ago | |
dec_model.png | 2 years ago | |
idec_model.png | 2 years ago |
Keras implementation for our IJCAI-17 paper:
and re-implementation for paper:
This code requires pretrained autoencoder weights provided.
Use IDEC-toy code for a quick start.
Install Keras v2.0, scikit-learn and git
sudo pip install keras scikit-learn
sudo apt-get install git
Clone the code to local.
git clone https://github.com/XifengGuo/IDEC.git IDEC
Prepare datasets.
cd IDEC/data/usps
bash ./download_usps.sh
cd ../reuters
bash ./get_data.sh
cd ../..
Get pre-trained autoencoder's weights.
Follow instructions at https://github.com/piiswrong/dec to pre-train the autoencoder.
Then save the trained weights to a keras model (e.g. mnist_ae_weights.h5) and put it in folder 'ae_weights'.
If you do not want to install Caffe package, you can download the pretrained weights from
https://github.com/XifengGuo/data-and-models
Then put .h5 file in ae_weights in local folder 'ae_weights'.
Or you can just use IDEC-toy code for a quick start, but the results may be not promising.
Run experiment on MNIST.
python IDEC.py mnist --ae_weights ae_weights/mnist_ae_weights.h5
or
python DEC.py mnist --ae_weights ae_weights/mnist_ae_weights.h5
The IDEC (or DEC) model is saved to "results/idec/IDEC_model_final.h5" (or "results/dec/DEC_model_final.h5").
Run experiment on USPS.
python IDEC.py usps --ae_weights ae_weights/usps_ae_weights --update_interval 30
python DEC.py usps --ae_weights ae_weights/usps_ae_weights --update_interval 30
Run experiment on REUTERSIDF10K.
python IDEC.py reutersidf10k --ae_weights ae_weights/reutersidf10k_ae_weights --n_clusters 4 --update_interval 3
python DEC.py reutersidf10k --ae_weights ae_weights/reutersidf10k_ae_weights --n_clusters 4 --update_interval 20
No Description
Python Shell
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》