issues
search
torchmd
/
torchmd-net
Training neural network potentials
MIT License
335
stars
75
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Prior: D2 dispersive correction
#143
raimis
closed
2 years ago
2
Support atom-wise and molecule-wise priors
#142
raimis
closed
2 years ago
8
POC: Enable to train on the partial charges
#141
raimis
closed
11 months ago
1
Allow the Ace loader to load multiple dataset at once
#140
raimis
closed
2 years ago
0
Implement Ace 2 loader
#139
raimis
closed
2 years ago
0
Fix the sign of the Ace loader
#138
raimis
closed
2 years ago
1
Catch null dataset_arg
#137
PhilippThoelke
closed
2 years ago
0
Improve the speed of the Ace loader
#136
raimis
closed
2 years ago
1
Access attribute instead of parameter
#135
PhilippThoelke
closed
2 years ago
0
Implement ZBL potential
#134
peastman
closed
2 years ago
10
Move the reduction into the output model
#133
raimis
closed
2 years ago
1
Implement mol_id for the loaders
#132
raimis
closed
2 years ago
2
Add W&B
#131
giadefa
closed
1 year ago
1
Rename cli
#130
giadefa
closed
2 years ago
0
SPICE example
#129
raimis
closed
2 years ago
0
Implement molecule sub-sampling
#128
raimis
closed
2 years ago
1
Implement CLI
#127
raimis
closed
2 years ago
2
Allow to select SPICE version
#126
raimis
closed
2 years ago
1
bad_alloc using PyTorch 1.12
#125
peastman
closed
10 months ago
7
Update the Ace loader
#124
raimis
closed
2 years ago
1
Reproduding QM9 Results
#123
YZHANG1996
closed
2 years ago
3
Cannot create env with mamba
#122
geemi725
closed
2 years ago
10
Consistent naming between dy and forces
#121
PhilippThoelke
closed
2 years ago
11
Test loss
#120
raimis
opened
2 years ago
2
Fix the gradient sign for QM9q and SPICE
#119
raimis
closed
2 years ago
1
Fix the usage of gradients
#118
raimis
closed
2 years ago
5
Uniform the usages of gradients
#117
raimis
closed
2 years ago
1
Inconsistency between gradients and forces
#116
raimis
closed
2 years ago
10
Refactor the additional argument passing to dataset classes
#115
stefdoerr
closed
2 years ago
0
QM9q: load dipoles and partial charges
#114
raimis
closed
2 years ago
1
Improved SPICE loader
#113
raimis
closed
2 years ago
2
Update the installation instructions
#112
raimis
closed
2 years ago
4
Problems when run example QM9
#111
limin2021
closed
2 years ago
4
Bug? MD17 stores all raw paths to the same processed_path
#110
basveeling
closed
2 years ago
0
auto-formatting with black
#109
stefdoerr
closed
2 years ago
0
Problem creating an environment
#108
limin2021
closed
2 years ago
5
Formation energies
#107
stefdoerr
closed
2 years ago
0
Comp6 formation energies
#106
stefdoerr
closed
2 years ago
3
CI: Use Mamba
#105
raimis
closed
2 years ago
0
Fix SPICE loader
#104
raimis
closed
2 years ago
3
Fix the QM9q loader
#103
raimis
closed
2 years ago
1
QM9q: implement the reference energy computation
#102
raimis
closed
2 years ago
1
Dataloader for SPICE
#101
raimis
closed
2 years ago
0
Fix the force unit in QM9q
#100
raimis
closed
2 years ago
0
Loader for QM9q dataset
#99
raimis
closed
2 years ago
0
reduce the dependencies of torchmd-net
#98
stefdoerr
closed
2 years ago
0
Attempt at upgrading pytorch-lightning
#97
PhilippThoelke
closed
2 years ago
3
Ways to reduce memory use
#96
peastman
opened
2 years ago
15
Question about adding missing atoms in protein
#95
psp3dcg
closed
2 years ago
10
Handle when load_model is specified twice
#94
PhilippThoelke
closed
2 years ago
0
Previous
Next