Learning Molecular Dynamics with Simple Language Model built upon Long Short-Term Memory (LSTM) Neural Network
This is an example showing how to use a simple LSTM based lagnuage model to learn a Langevin dynamics of a 4-state model potential described in Nat. Commu. 11, 5115 (2020). Please read and cite these manuscripts if using this example: Nat. Commu. 11, 5115 (2020)
This file contains all settings, data preprocessing, loss function, and the model itself. The program is tested under tensorflow/1.10.1 with keras beckend. Simpy implement this file by
python3 LSTM_LM_4s.py
in the terminal. The program will save a file with prediction and corresponding training checkpoints of last few epochs.
This program uses this file as input loss function. For using other loss, removing line 179 and return in line 182:
from lossT import sparse_categorical_crossentropy
def loss(labels, logits):
return sparse_categorical_crossentropy(labels, logits, from_logits=True)
Portions of this page are modifications based on work created and shared by Google and used according to terms described in the Creative Commons 4.0 Attribution License.