This is the implementation of Variational Autoencoder Inverse Mapper: An End-to-End Deep Learning Framework for Inverse Problems (VAIM). (https://ieeexplore.ieee.org/document/9534012)
The code is written in Python=3.6, with the following libraries:
- tensorflow==1.11.0
- keras==2.1.2
- Install the python libraries. (See Requirements).
- Download the code from GitHub:
git clone https://github.com/alanaziyasir/VAIM
cd VAIM- Run the python script:
python3 train.py- By default the script will run the first toy example which is f(x) = x2.
- To run another example, adjust self.example variable in line 12 in VAIM.py.
- To see the jupyter notebbok demo go to VAIM_demo.ipynb.
- The script will create a directory saved_model/ and save the the weights with the lowest validation error
- It will also plot the latent and the results
| f(x) = x2 | f(x) = sin(x) |
|---|---|
 |
 |
| latent of x2 | latent of sin(x) |
|---|---|
 |
 |