We provide the implementation and experiment results for the paper: Bayesian Convolutional Deep Sets with Task-Dependent Stationary Prior
- examples_construct_representation.ipynb : examples of random functional representation
- examples_task1_dataset.ipynb : examples of tasks in section 4.1
- examples_task1_singlechannel_regression.ipynb : experiment results in section 4.1
- examples_task2_multichannel_regression.ipynb : experiment results in section 4.2
- examples_task3_imagecompletion.ipynb : experiment results in section 4.3
- examples_task4_predatorpray.ipynb : experiment results in Appendix E
- python >= 3.6
- torch = 1.7
- pandas
- scipy
- attrdict
if necessary, install the required module as follows
pip3 install module-name
ex) pip3 install numpy