Comments (7)
from mace.
One can now specify --loss_fn="forces_only"
to train on data without energies. We still need to find a solution for mixed missing data, probably some masking scheme.
from mace.
The models should now be able to handle missing labels on specific config type since #22 and #26. @davkovacs @bernstei How important is to support missing labels within the same config types. If it is important, how is it handled in GAP? It would be a per config weighting?
from mace.
What do you mean by missing "labels"? Missing quantities like energy/forces/virial ?
from mace.
Yes exactly.
from mace.
Per-config weights are supported, and that's how missing data is handled. In GAP I believe that the design matrix construction is not batched, so for every config rows are either added or not depending on what quantities are available.
Per-config weights are done in config_from_atoms()
in https://github.com/ACEsuit/mace/blob/develop/mace/data/utils.py. For example line 133 reads a weight from Atoms.info["config_energy_weight"]
, or sets the weight to 0 if the energy is missing in line 139
from mace.
Perfect, so I think we need to add that to the readme, and close this issue.
from mace.
Related Issues (20)
- How do you make mutli-GPU training? HOT 2
- a couple of weird statements in new multihead fine-tuning interface code HOT 2
- training with foudational_model HOT 12
- How to run multi-GPU with mace-lammps?
- Error reading 'extended' xyz format files HOT 23
- What are the node energies? HOT 2
- Training flag not working HOT 2
- Dipole fitting on multi-GPU
- Is 0.694 timesteps/s a normal speed?
- select pretrained configs for multihead replay with or without replacement? HOT 2
- adding cmd argument for specifying wandb directory HOT 2
- CUDA Out of Memory with GCMC
- Distributed Training: invalid device ordinal
- Missing dependancies in `multi-head-interface` HOT 1
- Stress / num_atoms in loss? HOT 3
- stress errors reported during fitting and at end should be consistent HOT 4
- during-training valid loss is wrong HOT 11
- np.random.choice for selecting pre-fit subset for multihead fine tune doesn't accept a list of Atoms
- multi-head-finetune validation batches not deterministic HOT 3
- installation HOT 6
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
π Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. πππ
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google β€οΈ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from mace.