txie-93 / gdynet Goto Github PK
View Code? Open in Web Editor NEWUnsupervised learning of atomic scale dynamics from molecular dynamics.
License: MIT License
Unsupervised learning of atomic scale dynamics from molecular dynamics.
License: MIT License
def _pooling(inps):
"""
Pool the atom_fea of target atoms together using their indexes.
Parameters
----------
atom_fea: (B, N, atom_fea_len)
target_index: (B, N0)
Returns
-------
crys_fea: (B, N0, atom_fea_len)
"""
atom_fea, target_index = inps
B = tf.shape(atom_fea)[0]
N0 = tf.shape(target_index)[1]
batch_idx = tf.reshape(tf.range(0, B), (B, 1))
batch_idx = tf.tile(batch_idx, (1, N0))
full_idx = tf.stack((batch_idx, target_index), axis=-1)
return tf.gather_nd(atom_fea, full_idx)
dear doc Xie:
I am trying to use gdynet to identify dynamic processes in a friction MD simulation to get specific bonding state for aimed atom. After training the model , I follow the jupyter notebook to visualize the results. When I try to plot the relaxation timescales, something wrong happens just like:
gdynet/postprocess.py:41: RuntimeWarning: invalid value encountered in log
its_log_std = np.std(np.log(splited_its), axis=0)
/conda/anaconda3/envs/gdynet/lib/python3.6/site-packages/numpy/core/_methods.py:117: RuntimeWarning: invalid value encountered in subtract
x = asanyarray(arr - arrmean)
So I check the koopman_op for the array lags, finding there are a lot of zero arrays (the second one):
[[ 6.7463994e-01 -1.1839467e-01 4.3136746e-01 1.2387081e-02]
[-5.4880998e-06 9.7734249e-01 2.0228788e-02 2.4341424e-03]
[ 2.2197758e-04 9.0464517e-02 9.0913689e-01 1.7680590e-04]
[ 2.5602803e-05 7.5335771e-02 -1.7902499e-02 9.4254106e-01]]
[[0. 0. 0. 0.]
[0. 0. 0. 0.]
[0. 0. 0. 0.]
[0. 0. 0. 0.]]
I am wondering which part goes wrong. Thanks a lot!
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.