Datasets:
SinaMostafanejad
commited on
Commit
·
1e7b005
1
Parent(s):
acc1f58
adds the dataset card
Browse files
README.md
CHANGED
@@ -1,87 +1,279 @@
|
|
1 |
-
|
2 |
-
|
3 |
-
|
4 |
-
|
5 |
-
|
6 |
-
|
7 |
-
-
|
8 |
-
|
9 |
-
-
|
10 |
-
|
11 |
-
-
|
12 |
-
|
13 |
-
-
|
14 |
-
|
15 |
-
|
16 |
-
|
17 |
-
|
18 |
-
|
19 |
-
|
20 |
-
|
21 |
-
-
|
22 |
-
|
23 |
-
-
|
24 |
-
dtype: string
|
25 |
-
- name: oechem-version
|
26 |
-
dtype: string
|
27 |
-
- name: mopac-keywords
|
28 |
-
sequence: string
|
29 |
-
- name: description
|
30 |
-
dtype: string
|
31 |
-
- name: atomic-indices
|
32 |
-
sequence: int64
|
33 |
-
- name: atomic-numbers
|
34 |
-
sequence: int64
|
35 |
-
- name: atomic-symbols
|
36 |
-
sequence: string
|
37 |
-
- name: coordinates
|
38 |
-
sequence: float64
|
39 |
-
- name: bonds
|
40 |
-
sequence: int64
|
41 |
-
- name: enthalpy-of-formation
|
42 |
-
dtype: float64
|
43 |
-
- name: enthalpy-of-formation-error
|
44 |
-
dtype: string
|
45 |
-
- name: enthalpy-of-formation-reference
|
46 |
-
dtype: 'null'
|
47 |
-
- name: enthalpy-of-formation-units
|
48 |
-
dtype: string
|
49 |
-
- name: ionization-energy
|
50 |
-
dtype: float64
|
51 |
-
- name: ionization-energy-error
|
52 |
-
dtype: string
|
53 |
-
- name: ionization-energy-reference
|
54 |
-
dtype: 'null'
|
55 |
-
- name: ionization-energy-units
|
56 |
-
dtype: string
|
57 |
-
- name: entropy
|
58 |
-
dtype: float64
|
59 |
-
- name: entropy-units
|
60 |
-
dtype: string
|
61 |
-
- name: constant-pressure-heat-capacity
|
62 |
-
dtype: float64
|
63 |
-
- name: constant-pressure-heat-capacity-units
|
64 |
-
dtype: string
|
65 |
-
- name: diople-moment
|
66 |
-
dtype: float64
|
67 |
-
- name: dipole-moment-reference
|
68 |
-
dtype: string
|
69 |
-
- name: dipole-moment-units
|
70 |
-
dtype: string
|
71 |
-
- name: mopac-reference-energy
|
72 |
-
dtype: float64
|
73 |
-
- name: mopac-reference-energy-units
|
74 |
-
dtype: string
|
75 |
-
splits:
|
76 |
-
- name: train
|
77 |
-
num_bytes: 6783360
|
78 |
-
num_examples: 4617
|
79 |
-
download_size: 2172928
|
80 |
-
dataset_size: 6783360
|
81 |
configs:
|
82 |
-
- config_name: pm7
|
83 |
-
|
84 |
-
|
85 |
-
|
86 |
-
|
87 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
--
|
2 |
+
license:
|
3 |
+
- lgpl-3.0
|
4 |
+
license_link: https://www.gnu.org/licenses/lgpl-3.0.en.html
|
5 |
+
tags:
|
6 |
+
- heat-of-formation
|
7 |
+
- dipole-moments
|
8 |
+
- entropy
|
9 |
+
- energy
|
10 |
+
- pm7
|
11 |
+
- crystals
|
12 |
+
- small-molecules
|
13 |
+
- organometallics
|
14 |
+
pretty_name: pm7
|
15 |
+
size_categories:
|
16 |
+
- n<10K
|
17 |
+
source_datasets:
|
18 |
+
- pm7
|
19 |
+
task_categories:
|
20 |
+
- tabular-regression
|
21 |
+
- other
|
22 |
+
task_ids:
|
23 |
+
- tabular-single-column-regression
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
24 |
configs:
|
25 |
+
- config_name: pm7
|
26 |
+
data_files:
|
27 |
+
- split: train
|
28 |
+
path: "data/train*.parquet"
|
29 |
+
default: true
|
30 |
---
|
31 |
+
# PM7 Dataset
|
32 |
+
|
33 |
+
## Table of Contents
|
34 |
+
|
35 |
+
- [PM7 Dataset](#pm7-dataset)
|
36 |
+
- [Table of Contents](#table-of-contents)
|
37 |
+
- [Dataset Description](#dataset-description)
|
38 |
+
- [Dataset Summary](#dataset-summary)
|
39 |
+
- [Dataset Structure](#dataset-structure)
|
40 |
+
- [Data Instances](#data-instances)
|
41 |
+
- [Data Fields](#data-fields)
|
42 |
+
- [Data Splits and Configurations](#data-splits-and-configurations)
|
43 |
+
- [How to Use the Dataset](#how-to-use-the-dataset)
|
44 |
+
- [Prerequisites](#prerequisites)
|
45 |
+
- [Accessing the Data](#accessing-the-data)
|
46 |
+
- [Dataset Creation](#dataset-creation)
|
47 |
+
- [Curation Rationale](#curation-rationale)
|
48 |
+
- [Initial Data Collection and Normalization](#initial-data-collection-and-normalization)
|
49 |
+
- [Personal and Sensitive Information](#personal-and-sensitive-information)
|
50 |
+
- [Considerations for Using the Data](#considerations-for-using-the-data)
|
51 |
+
- [Social Impact of Dataset](#social-impact-of-dataset)
|
52 |
+
- [Additional Information](#additional-information)
|
53 |
+
- [Dataset Curators](#dataset-curators)
|
54 |
+
- [Licensing Information](#licensing-information)
|
55 |
+
- [Citation Information](#citation-information)
|
56 |
+
- [Contributions](#contributions)
|
57 |
+
|
58 |
+
## Dataset Description
|
59 |
+
|
60 |
+
- **Homepage:** http://openmopac.net
|
61 |
+
- **Repository:** https://github.com/openmopac/mopac
|
62 |
+
- **Paper:** https://doi.org/10.1007/s00894-012-1667-x
|
63 |
+
- **Point of Contact:** [James J. P. Stewart]([email protected])
|
64 |
+
- **Point of Contact:** [Paul Saxe]([email protected])
|
65 |
+
- **Point of Contact:** [Mohammad Mostafanejad]([email protected])
|
66 |
+
- **Point of Contact:** [MolSSI-AI Hub]([email protected])
|
67 |
+
|
68 |
+
### Dataset Summary
|
69 |
+
|
70 |
+
PM7 is a modern semi-empirical method with broad applications in chemical space
|
71 |
+
including organic molecules, organometallics and solid-state crystalline
|
72 |
+
compounds. The parameterization process involved adopting a diverse set of
|
73 |
+
experimental and high-level ab initio reference data. The current dataset,
|
74 |
+
curated by the Molecular Sciences Software Institute, attempts to provide a
|
75 |
+
ready-to-use version of the PM7 dataset for the computational molecular science
|
76 |
+
and machine learning community.
|
77 |
+
|
78 |
+
## Dataset Structure
|
79 |
+
|
80 |
+
### Data Instances
|
81 |
+
|
82 |
+
An example of a data instance is as follows:
|
83 |
+
|
84 |
+
```json
|
85 |
+
{'title': 'MOPAC_54/PM7_reference',
|
86 |
+
'name': '1,1,1-Trifluoroethane',
|
87 |
+
'mopac-id': 54,
|
88 |
+
'net-charge': 0,
|
89 |
+
'spin-multiplicity': 1,
|
90 |
+
'openbabel-canonical-smiles': 'CC(F)(F)F',
|
91 |
+
'openbabel-version': '3.1.0',
|
92 |
+
'rdkit-canonical-smiles': 'CC(F)(F)F',
|
93 |
+
'rdkit-version': '2024.03.5',
|
94 |
+
'oechem-canonical-smiles': 'CC(F)(F)F',
|
95 |
+
'oechem-version': '20240905',
|
96 |
+
'mopac-keywords': ['SYMMETRY', 'PM7'],
|
97 |
+
'description': '1,1,1-Trifluoroethane\nI=13.8 IR=LLNBS82 D=2.32 H=-178.9 HR=NIST DR=MCC1963 S=66.9 CP=18.69',
|
98 |
+
'atomic-indices': [1, 2, 3, 4, 5, 6, 7, 8],
|
99 |
+
'atomic-numbers': [6, 6, 9, 9, 9, 1, 1, 1],
|
100 |
+
'atomic-symbols': ['C', 'C', 'F', 'F', 'F', 'H', 'H', 'H'],
|
101 |
+
'coordinates': [0.0,
|
102 |
+
0.0,
|
103 |
+
0.0,
|
104 |
+
...,
|
105 |
+
-0.41029998660087585,
|
106 |
+
-0.8860999941825867,
|
107 |
+
0.5116000175476074],
|
108 |
+
'bonds': [1, 7, 1, 1, 2, 1, 1, 6, 1, 1, 8, 1, 2, 5, 1, 2, 4, 1, 2, 3, 1],
|
109 |
+
'enthalpy-of-formation': -748.5176,
|
110 |
+
'enthalpy-of-formation-error': None,
|
111 |
+
'enthalpy-of-formation-reference': None,
|
112 |
+
'enthalpy-of-formation-units': 'kJ/mol',
|
113 |
+
'ionization-energy': 13.8,
|
114 |
+
'ionization-energy-error': None,
|
115 |
+
'ionization-energy-reference': None,
|
116 |
+
'ionization-energy-units': 'eV',
|
117 |
+
'entropy': 279.9096,
|
118 |
+
'entropy-units': 'J/mol/K',
|
119 |
+
'constant-pressure-heat-capacity': 78.19896000000001,
|
120 |
+
'constant-pressure-heat-capacity-units': 'J/mol/K',
|
121 |
+
'diople-moment': 2.32,
|
122 |
+
'dipole-moment-reference': 'MCC1963',
|
123 |
+
'dipole-moment-units': 'D',
|
124 |
+
'mopac-reference-energy': -731.158184,
|
125 |
+
'mopac-reference-energy-units': 'kJ/mol'}
|
126 |
+
```
|
127 |
+
|
128 |
+
### Data Fields
|
129 |
+
|
130 |
+
| Field | Description |
|
131 |
+
| ------------------------------------- | --------------------------------------------------------------- |
|
132 |
+
| title | Title line in the MOPAC input/SDF file |
|
133 |
+
| name | Name of the molecule |
|
134 |
+
| mopac-id | MOPAC ID of the molecule |
|
135 |
+
| net-charge | Net charge of the molecule |
|
136 |
+
| spin-multiplicity | Spin multiplicity of the molecule |
|
137 |
+
| openbabel-canonical-smiles | Canonical SMILES string generated by OpenBabel |
|
138 |
+
| openbabel-version | Version of OpenBabel used to generate the canonical SMILES |
|
139 |
+
| rdkit-canonical-smiles | Canonical SMILES string generated by RDKit |
|
140 |
+
| rdkit-version | Version of RDKit used to generate the canonical SMILES |
|
141 |
+
| oechem-canonical-smiles | Canonical SMILES string generated by OpenEye OEChem |
|
142 |
+
| oechem-version | Version of OpenEye OEChem used to generate the canonical SMILES |
|
143 |
+
| mopac-keywords | Keywords used in the MOPAC input file |
|
144 |
+
| description | Description of the molecule |
|
145 |
+
| atomic-indices | Atomic indices of the atoms in the molecule |
|
146 |
+
| atomic-numbers | Atomic numbers of the atoms in the molecule |
|
147 |
+
| atomic-symbols | Atomic symbols of the atoms in the molecule |
|
148 |
+
| coordinates | Atomic coordinates of the atoms in the molecule |
|
149 |
+
| bonds | Bond information of the molecule |
|
150 |
+
| enthalpy-of-formation | Enthalpy of formation of the molecule |
|
151 |
+
| enthalpy-of-formation-error | Error in the enthalpy of formation |
|
152 |
+
| enthalpy-of-formation-reference | Reference for the enthalpy of formation |
|
153 |
+
| enthalpy-of-formation-units | Units of the enthalpy of formation |
|
154 |
+
| ionization-energy | Ionization energy of the molecule |
|
155 |
+
| ionization-energy-error | Error in the ionization energy |
|
156 |
+
| ionization-energy-reference | Reference for the ionization energy |
|
157 |
+
| ionization-energy-units | Units of the ionization energy |
|
158 |
+
| entropy | Entropy of the molecule |
|
159 |
+
| entropy-units | Units of the entropy |
|
160 |
+
| constant-pressure-heat-capacity | Constant pressure heat capacity of the molecule |
|
161 |
+
| constant-pressure-heat-capacity-units | Units of the constant pressure heat capacity |
|
162 |
+
| diople-moment | Dipole moment of the molecule |
|
163 |
+
| dipole-moment-reference | Reference for the dipole moment |
|
164 |
+
| dipole-moment-units | Units of the dipole moment |
|
165 |
+
| mopac-reference-energy | Reference energy of the molecule |
|
166 |
+
| mopac-reference-energy-units | Units of the reference energy |
|
167 |
+
|
168 |
+
### Data Splits and Configurations
|
169 |
+
|
170 |
+
The dataset has only one `train` split. Currently, the PM7 dataset has one
|
171 |
+
configuration/subset:
|
172 |
+
|
173 |
+
- `pm7` (default)
|
174 |
+
|
175 |
+
## How to Use the Dataset
|
176 |
+
|
177 |
+
### Prerequisites
|
178 |
+
|
179 |
+
We recommend isolating your work in a virtualenv or conda environment.
|
180 |
+
You can create a new conda environment, `pm7`,
|
181 |
+
|
182 |
+
```bash
|
183 |
+
conda create -n pm7 python=3.12
|
184 |
+
```
|
185 |
+
|
186 |
+
and activate it using the following command
|
187 |
+
|
188 |
+
```bash
|
189 |
+
conda activate pm7
|
190 |
+
```
|
191 |
+
|
192 |
+
Once the conda environment is activated, you can
|
193 |
+
install the dependencies in it as shown below
|
194 |
+
|
195 |
+
```bash
|
196 |
+
pip install huggingface_hub
|
197 |
+
```
|
198 |
+
|
199 |
+
### Accessing the Data
|
200 |
+
|
201 |
+
Once the required packages are installed, you can run the following code
|
202 |
+
to access the data
|
203 |
+
|
204 |
+
```python
|
205 |
+
# import the modules
|
206 |
+
from datasets import load_dataset
|
207 |
+
# load the dataset with streaming
|
208 |
+
hub_ds = load_dataset(path="molssiai-hub/pm7",
|
209 |
+
name="pm7",
|
210 |
+
split="train",
|
211 |
+
cache_dir="./tmp")
|
212 |
+
# fetch a random batch of 32 samples from the dataset
|
213 |
+
ds = hub_ds.shuffle(seed=1234).select(range(32))
|
214 |
+
```
|
215 |
+
|
216 |
+
The argument `name` by default is set to `pm7`. The `split` must be set to
|
217 |
+
`train` as it is the only split in our dataset. The `cache_dir` allows us to
|
218 |
+
store the Hugging Face datasets' and models' artifacts in a non-default
|
219 |
+
directory (by default, it is set to `~/.cache/huggingface`). The `shuffle`
|
220 |
+
method shuffles the dataset with a seed of 1234, and the `select` method selects
|
221 |
+
the first 32 samples from the shuffled dataset.
|
222 |
+
|
223 |
+
## Dataset Creation
|
224 |
+
|
225 |
+
### Curation Rationale
|
226 |
+
|
227 |
+
The present version of PM7 dataset has been extracted from the
|
228 |
+
[pm7.sdf.gz](https://huggingface.co/datasets/molssiai-hub/pm7/blob/main/pm7.sdf.gz)
|
229 |
+
file, transformed into a dictionary and stored in the `.json` format.
|
230 |
+
|
231 |
+
#### Initial Data Collection and Normalization
|
232 |
+
|
233 |
+
Other than the changes detailed in Sec. [Curation Rationale](#curation-rationale),
|
234 |
+
no data modification has been performed on the PM7 dataset.
|
235 |
+
|
236 |
+
### Personal and Sensitive Information
|
237 |
+
|
238 |
+
The PM7 dataset does not involve any personal or sensitive information.
|
239 |
+
|
240 |
+
## Considerations for Using the Data
|
241 |
+
|
242 |
+
### Social Impact of Dataset
|
243 |
+
|
244 |
+
The PM7 dataset paves the way for applications in (in)organic chemistry,
|
245 |
+
solid-state physics and materials science, among others.
|
246 |
+
|
247 |
+
## Additional Information
|
248 |
+
|
249 |
+
### Dataset Curators
|
250 |
+
|
251 |
+
- **James J. P. Stewart**, Stewart Computational Chemistry, 15210 Paddington Circle, Colorado Springs, CO, 80921, USA
|
252 |
+
- **Paul Saxe**, The Molecular Sciences Software Institute (MolSSI), Blacksburg, VA, 24060, USA
|
253 |
+
- **Mohammad Mostafanejad**, The Molecular Sciences Software Institute (MolSSI), Blacksburg, VA, 24060, USA
|
254 |
+
|
255 |
+
### Licensing Information
|
256 |
+
|
257 |
+
[GNU lesser general public license version 3.0](https://www.gnu.org/licenses/lgpl-3.0.en.htm)
|
258 |
+
|
259 |
+
### Citation Information
|
260 |
+
|
261 |
+
```tex
|
262 |
+
@article{Stewart:2013:1,
|
263 |
+
author = {Stewart, James J. P.},
|
264 |
+
doi = {10.1007/S00894-012-1667-X},
|
265 |
+
journal = {Journal of Molecular Modeling},
|
266 |
+
pages = {1-32},
|
267 |
+
publisher = {Springer},
|
268 |
+
title = {Optimization of parameters for semiempirical methods VI: More modifications to the NDDO approximations and re-optimization of parameters},
|
269 |
+
volume = {19},
|
270 |
+
url = {https://link.springer.com/article/10.1007/s00894-012-1667-x},
|
271 |
+
year = {2013}
|
272 |
+
}
|
273 |
+
```
|
274 |
+
|
275 |
+
### Contributions
|
276 |
+
|
277 |
+
- **James J. P. Stewart**, Stewart Computational Chemistry
|
278 |
+
- **Paul Saxe**, The Molecular Sciences Software Institute (MolSSI)
|
279 |
+
- **Mohammad Mostafanejad**, The Molecular Sciences Software Institute (MolSSI)
|