Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -11,16 +11,24 @@ size_categories:
|
|
| 11 |
tags:
|
| 12 |
- drug discovery
|
| 13 |
- bioassay
|
| 14 |
-
dataset_summary:
|
| 15 |
-
|
|
|
|
| 16 |
drug discovery.
|
| 17 |
-
citation:
|
| 18 |
-
|
| 19 |
-
|
| 20 |
-
|
| 21 |
-
|
| 22 |
-
|
| 23 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 24 |
dataset_info:
|
| 25 |
config_name: MolData
|
| 26 |
features:
|
|
@@ -32,7 +40,7 @@ dataset_info:
|
|
| 32 |
dtype: string
|
| 33 |
- name: AID
|
| 34 |
dtype: string
|
| 35 |
-
- name: Y
|
| 36 |
dtype: int64
|
| 37 |
splits:
|
| 38 |
- name: train
|
|
@@ -151,4 +159,20 @@ Split and evaluate the catboost model
|
|
| 151 |
|
| 152 |
scores = regression_suite.compute(
|
| 153 |
references=split_featurised_dataset["test"]['Y'],
|
| 154 |
-
predictions=preds["cat_boost_regressor::Y"])
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
tags:
|
| 12 |
- drug discovery
|
| 13 |
- bioassay
|
| 14 |
+
dataset_summary: >-
|
| 15 |
+
A comprehensive disease and target-based dataset with 1.4 million molecules,
|
| 16 |
+
collected from PubChem to accelerate molecular machine learning for better
|
| 17 |
drug discovery.
|
| 18 |
+
citation: |-
|
| 19 |
+
@article{KeshavarziArshadi2022,
|
| 20 |
+
title = {MolData, a molecular benchmark for disease and target based machine learning},
|
| 21 |
+
volume = {14},
|
| 22 |
+
ISSN = {1758-2946},
|
| 23 |
+
url = {http://dx.doi.org/10.1186/s13321-022-00590-y},
|
| 24 |
+
DOI = {10.1186/s13321-022-00590-y},
|
| 25 |
+
number = {1},
|
| 26 |
+
journal = {Journal of Cheminformatics},
|
| 27 |
+
publisher = {Springer Science and Business Media LLC},
|
| 28 |
+
author = {Keshavarzi Arshadi, Arash and Salem, Milad and Firouzbakht, Arash and Yuan, Jiann Shiun},
|
| 29 |
+
year = {2022},
|
| 30 |
+
month = mar
|
| 31 |
+
}
|
| 32 |
dataset_info:
|
| 33 |
config_name: MolData
|
| 34 |
features:
|
|
|
|
| 40 |
dtype: string
|
| 41 |
- name: AID
|
| 42 |
dtype: string
|
| 43 |
+
- name: 'Y'
|
| 44 |
dtype: int64
|
| 45 |
splits:
|
| 46 |
- name: train
|
|
|
|
| 159 |
|
| 160 |
scores = regression_suite.compute(
|
| 161 |
references=split_featurised_dataset["test"]['Y'],
|
| 162 |
+
predictions=preds["cat_boost_regressor::Y"])
|
| 163 |
+
|
| 164 |
+
|
| 165 |
+
### Citation
|
| 166 |
+
@article{KeshavarziArshadi2022,
|
| 167 |
+
title = {MolData, a molecular benchmark for disease and target based machine learning},
|
| 168 |
+
volume = {14},
|
| 169 |
+
ISSN = {1758-2946},
|
| 170 |
+
url = {http://dx.doi.org/10.1186/s13321-022-00590-y},
|
| 171 |
+
DOI = {10.1186/s13321-022-00590-y},
|
| 172 |
+
number = {1},
|
| 173 |
+
journal = {Journal of Cheminformatics},
|
| 174 |
+
publisher = {Springer Science and Business Media LLC},
|
| 175 |
+
author = {Keshavarzi Arshadi, Arash and Salem, Milad and Firouzbakht, Arash and Yuan, Jiann Shiun},
|
| 176 |
+
year = {2022},
|
| 177 |
+
month = mar
|
| 178 |
+
}
|