Operator Learning for Bubble Growth Dynamics

Sayan - LSTM + GRU analysis Tarun - Seq2Seq analysis Vivek - DeepONet analysis

Each folder has arch_m-value folders with respective analysis with varying m values.

For plots, each arch_m-value folder (eg: gru_20) has predictions folder which has plots for different l values (length scale of the Gaussian Random Field).

To run DeepOnet:

    python train_don.py

To run Seq2Seq:

    python train_seq.py

To run LSTM/GRU:

    python train.py LSTM 20
    python train.py GRU 20

To run analysis on the trained data:

For DeepONet

    jupyter analyze.ipynb

For Seq2Seq

    python analyze_seq.py

For LSTM/GRU

    python analyze.py LSTM 20
    python analyze.py GRU 20