Two Submissions on Clearance

README.md

@@ -24,3 +24,10 @@
 | ---- | ------------- | -------------- | ---------------- | ------------------ | --------------------------------- | ---------------------------------------------------------------------------------------- | -------------- |
 | 1    | Random Forest | 1024-bit ECFP4 | 0.9540 +- 0.0038 | 0.9062 +- 0.0079   | [Mufei Li](mufeili1996@gmail.com) | [Paper](https://www.stat.berkeley.edu/~breiman/randomforest2001.pdf), [Code](./examples) | Dec 30th, 2020 |
 | 2    | GCN           | GraphConv      | 0.9214 +- 0.0106 | 0.9445 +- 0.0049   | [Mufei Li](mufeili1996@gmail.com) | [Paper](https://arxiv.org/abs/1609.02907), [Code](./examples)                            | Dec 30th, 2020 |
+
+### Clearance
+
+| Rank | Model         | Featurization  | Test RMSE         | Validation RMSE   | Contact                           | References	                                                                           | Date           |
+| ---- | ------------- | -------------- | ----------------- | ----------------- | --------------------------------- | ---------------------------------------------------------------------------------------- | -------------- |
+| 1    | Random Forest | 1024-bit ECFP4 | 46.7056 +- 1.2794 | 43.1696 +- 0.9360 | [Mufei Li](mufeili1996@gmail.com) | [Paper](https://www.stat.berkeley.edu/~breiman/randomforest2001.pdf), [Code](./examples) | Jan 10th, 2021 |
+| 2    | GCN           | GraphConv      | 51.2271 +- 2.2749 | 42.1724 +- 1.8236 | [Mufei Li](mufeili1996@gmail.com) | [Paper](https://arxiv.org/abs/1609.02907), [Code](./examples)                            | Jan 10th, 2021 |

examples/README.md

@@ -18,6 +18,7 @@ The feasible arguments include:
         - `BACE_classification`
         - `BACE_regression`
         - `BBBP`
+        - `Clearance`
 - **Hyperparameter Search (optional)**: `-hs`
     - Perform a hyperparameter search using Bayesian optimization. It determines the best 
       hyperparameters based on the validation metric averaged across 3 runs.
@@ -40,6 +41,7 @@ The feasible arguments include:
         - `BACE_classification`
         - `BACE_regression`
         - `BBBP`
+        - `Clearance`
 - **Hyperparameter Search (optional)**: `-hs`
     - Perform a hyperparameter search using Bayesian optimization. It determines the best 
       hyperparameters based on the validation metric averaged across 3 runs.

examples/configures/GCN_GC/Clearance.json

@@ -0,0 +1,8 @@
+{
+  "batchnorm": false,
+  "dropout": 0.35007488294064143,
+  "hidden_feats": 64,
+  "lr": 0.2062498552628446,
+  "num_gnn_layers": 4,
+  "residual": false
+}

examples/configures/RF_ECFP/Clearance.json

@@ -0,0 +1,6 @@
+{
+  "bootstrap": true,
+  "criterion": "mse",
+  "min_samples_split": 4,
+  "n_estimators": 10
+}

examples/fingerprint.py

@@ -30,7 +30,7 @@ def rf_model_builder(model_dir, hyperparams, mode):
 def load_model(args, tasks, hyperparams):
   if args['dataset'] in ['BACE_classification', 'BBBP']:
     mode = 'classification'
-  elif args['dataset'] in ['BACE_regression']:
+  elif args['dataset'] in ['BACE_regression', 'Clearance']:
     mode = 'regression'
   else:
     raise ValueError('Unexpected dataset: {}'.format(args['dataset']))
@@ -154,7 +154,7 @@ def objective(hyperparams):
   parser.add_argument(
       '-d',
       '--dataset',
-      choices=['BACE_classification', 'BACE_regression', 'BBBP'],
+      choices=['BACE_classification', 'BACE_regression', 'BBBP', 'Clearance'],
       help='Dataset to use')
   parser.add_argument(
       '-m',

examples/gnn.py

@@ -15,7 +15,7 @@ def load_model(save_pth, args, tasks, hyperparams):
     mode = 'classification'
     # binary classification
     n_classes = 2
-  elif args['dataset'] in ['BACE_regression']:
+  elif args['dataset'] in ['BACE_regression', 'Clearance']:
     mode = 'regression'
     n_classes = None
   else:
@@ -177,7 +177,7 @@ def objective(hyperparams):
   parser.add_argument(
       '-d',
       '--dataset',
-      choices=['BACE_classification', 'BACE_regression', 'BBBP'],
+      choices=['BACE_classification', 'BACE_regression', 'BBBP', 'Clearance'],
       help='Dataset to use')
   parser.add_argument(
       '-m',

examples/utils.py

@@ -6,7 +6,7 @@
 def decide_metric(dataset):
   if dataset in ['BACE_classification', 'BBBP']:
     return 'roc_auc'
-  elif dataset == 'BACE_regression':
+  elif dataset in ['BACE_regression', 'Clearance']:
     return 'rmse'
   else:
     return ValueError('Unexpected dataset: {}'.format(dataset))
@@ -75,6 +75,10 @@ def load_dataset(args):
     from deepchem.molnet import load_bace_regression
     tasks, all_dataset, transformers = load_bace_regression(
         featurizer=featurizer, splitter=splitter, reload=False)
+  elif args['dataset'] == 'Clearance':
+    from deepchem.molnet import load_clearance
+    tasks, all_dataset, transformers = load_clearance(
+        featurizer=featurizer, splitter=splitter, reload=False)
   else:
     raise ValueError('Unexpected dataset: {}'.format(args['dataset']))