Comments (4)
Yes. I think we can close this issue
from nequip.
Hi @mkphuthi ,
In ASE this is unexpected... I don't get a segfault when trained with default_dtype: float64
added to minimal.yaml
and then running:
In [1]: from nequip.ase import NequIPCalculator
In [3]: calc = NequIPCalculator.from_deployed_model("test.pth", "cuda")
/home/a/Documents/NequIP/code/nequip/nequip/utils/_global_options.py:45: UserWarning: Setting the GLOBAL value for allow_tf32 to False which is different than the previous value of True
warnings.warn(
/home/a/Documents/NequIP/code/nequip/nequip/utils/_global_options.py:58: UserWarning: Setting the GLOBAL value for jit fusion strategy to `[('DYNAMIC', 3)]` which is different than the previous value of `[('STATIC', 2), ('DYNAMIC', 10)]`
warnings.warn(
/home/a/Documents/NequIP/code/nequip/nequip/ase/nequip_calculator.py:73: UserWarning: Trying to use chemical symbols as NequIP type names; this may not be correct for your model! To avoid this warning, please provide `species_to_type_name` explicitly.
warnings.warn(
In [4]: import ase.build
In [6]: atoms = ase.build.molecule("CH3CHO")
In [7]: atoms.calc = calc
In [8]: atoms.get_potential_energy()
---------------------------------------------------------------------------
RuntimeError Traceback (most recent call last)
...
RuntimeError: The following operation failed in the TorchScript interpreter.
Traceback of TorchScript, serialized code (most recent call last):
...
RuntimeError: expected scalar type Float but found Double
In [9]: import torch
In [10]: torch.set_default_dtype(torch.float64)
In [12]: atoms.get_potential_energy()
Out[12]: -135216.08807821615
This could be called a bug (maybe it should detect automatically and convert to float64, or at least try to set the global set_default_dtype
like the other global settings), but it certainly isn't a segfault... is this only one particular installation / environment where this occurs, or you've reproduced it in multiple places / ways?
In LAMMPS this is currently expected since we at present only support float32
inputs into the model...
from nequip.
Setting the default dtype to float64 fixed it! The segfault has been there with every NequIP install I have tried so far. Strange.
from nequip.
Weird... this should error out normally but at that point I'd attribute the problem to PyTorch. Is this all on the same machine, or across machines too that you see this?
Either way, glad that it is resolved. Can I close the issue then for the segfault?
from nequip.
Related Issues (20)
- β [QUESTION] Restart run HOT 1
- issue when using nequip-deploy π [BUG] HOT 9
- π [BUG] Cannot restart run with different dataset HOT 4
- Is it possible to train on xyz format data with multiple moleculesβ [QUESTION] HOT 2
- run works on colab but fails on spyderβ [QUESTION] HOT 4
- β [QUESTION] Custom layer with control structure not supported? HOT 1
- The use of nequip command HOT 1
- π [BUG] Cannot run nequip-train with provided example HOT 4
- π [FEATURE]How to Train and Validate on Separate Datasets HOT 2
- How to do custom EarlyStopping?β [QUESTION] HOT 4
- β [QUESTION]
- β [QUESTION] About the data class AtomicData HOT 3
- bugs with "initialize_from_state"π [BUG] HOT 5
- how to choose nosehoover the value of nvt_qβ [QUESTION]
- β [QUESTION]Finetuning Validation Error Higher than Pre-training Error in Nequip HOT 1
- Confusion about `num_frames` attribute in `HDF5Dataset` HOT 1
- β [QUESTION]how can I use the ase calculator for testing ?
- What is the unit of virialsβ [QUESTION] HOT 3
- MLFF for Silicon HOT 1
- minimal extxyz HOT 2
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 nequip.