File size: 2,076 Bytes
7b37970
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
---
library_name: pytorch
tags:
- molecular-generation
- flow-matching
- geom
- qm9
---
# Molecular LPGM

[Reference implementation and documentation](https://github.com/MurrellGroup/molLPGM)

Final **Fragment Noise** and **Brownian Atoms** models for GEOM-Drugs and QM9.
Both methods condition on atom count and the BRICS fragment partition, with
learned same/different-fragment attention biases in every transformer layer.
Neither receives clean coordinates, atom labels, bonds or geometric templates
at generation. Fragment Noise uses a single fixed global angular variance of
**6** in both datasets.

| Dataset | Width | Layers | Fragment Noise | Brownian Atoms |
| --- | ---: | ---: | --- | --- |
| GEOM-Drugs | 768 | 16 | [fragment.pt](geom/fragment.pt) | [bm.pt](geom/bm.pt) |
| QM9 | 512 | 12 | [fragment.pt](qm9/fragment.pt) | [bm.pt](qm9/bm.pt) |

All four checkpoints contain final **800,000-update raw and EMA weights**
(decay **0.9999**), architecture, process parameters, and final loss weights.
The final 100,000 updates are learning-rate warmdown. Optimizer state is omitted.
Use `--ema` in the reference sampling commands, or `load_checkpoint(...,
use_ema=True)` in Python. SHA-256 hashes and parameter counts are in
[checkpoints.json](checkpoints.json).

The source is a centered Gaussian with standard deviation 3 Å. Brownian Atoms
uses the scalar clock `u(t) = 1 - (1 - t**2.3)**3.1` and variance coefficient
`beta = 14.8 Ų`; these parameters approximately match the conditional
endpoint spread. It has one Cartesian coordinate head and the same atom-label
process as Fragment Noise.

[10K-sample results and archives](https://github.com/MurrellGroup/molLPGM/tree/main/results)
use final EMA weights and 2,000 square-root-spaced steps. The same ADiT validity,
valid-and-connected, and uniqueness protocol is used for both datasets.
GEOM evaluation uses test cases; QM9 pools validation and test cases.
See the [comparison protocol](https://github.com/MurrellGroup/molLPGM/blob/main/docs/comparison.md)
for conditioning, objective and training details.