update package version

README.md

@@ -15,14 +15,14 @@ The goal of model-based optimization is to find an input **x** that maximizes an
 Design-Bench can be installed with the complete set of benchmarks via our pip package.
 
 ```bash
-pip install design-bench[all]==2.0.14
+pip install design-bench[all]==2.0.15
 pip install morphing-agents==1.5.1
 ```
 
 Alternatively, if you do not have MuJoCo, you may opt for a minimal install.
 
 ```bash
-pip install design-bench==2.0.14
+pip install design-bench==2.0.15
 ```
 
 ## Available Tasks

example_iteration_script.py

@@ -0,0 +1,59 @@
+import pickle as pkl
+import design_bench as db
+import numpy as np
+
+
+if __name__ == "__main__":
+
+    with open("type_assay_pairs.pkl", "wb") as f:
+        type_assay_pairs = pkl.load(f)
+
+    all_rank_corr = []
+
+    for type_name, assay_name in type_assay_pairs:
+
+        task = db.make(
+            'ChEMBLMorganFingerprint-FullyConnected-v0',
+
+            dataset_kwargs=dict(
+                max_samples=None,
+                distribution=None,
+                max_percentile=50,
+                min_percentile=0,
+                assay_chembl_id=assay_name,
+                standard_type=type_name),
+
+            oracle_kwargs=dict(
+                noise_std=0.0,
+                max_samples=None,
+                distribution=None,
+                max_percentile=100,
+                min_percentile=0,
+
+                feature_extractor=MorganFingerprintFeatures(dtype=np.float32),
+
+                model_kwargs=dict(
+                    hidden_size=512,
+                    activation='relu',
+                    num_layers=2,
+                    epochs=5,
+                    shuffle_buffer=5000,
+                    learning_rate=0.0001),
+
+                split_kwargs=dict(
+                    val_fraction=0.1,
+                    subset=None,
+                    shard_size=50000,
+                    to_disk=True,
+                    disk_target=f"chembl-{type_name}-{assay_name}/split",
+                    is_absolute=False))
+
+        )
+
+        print(type_name, assay_name,
+              task.oracle.params['rank_correlation'])
+
+        all_rank_corr.append(task.oracle.params['rank_correlation'])
+
+    best_type_name, best_assay_name = \
+        type_assay_pairs[np.argmax(np.array(all_rank_corr))]

setup.py

@@ -7,7 +7,7 @@
     LONG_DESCRIPTION = readme.read()
 
 
-setup(name='design-bench', version='2.0.14', license='MIT',
+setup(name='design-bench', version='2.0.15', license='MIT',
       packages=find_packages(include=['design_bench', 'design_bench.*']),
       description='Design-Bench: Benchmarks for '
                   'Data-Driven Offline Model-Based Optimization',
@@ -17,7 +17,7 @@
       author_email='[email protected]',
       url='https://github.com/brandontrabucco/design-bench',
       download_url='https://github.com/'
-                   'brandontrabucco/design-bench/archive/v2_0_14.tar.gz',
+                   'brandontrabucco/design-bench/archive/v2_0_15.tar.gz',
       keywords=['Deep Learning', 'Neural Networks',
                 'Benchmark', 'Model-Based Optimization'],
       extras_require={'all': ['gym[mujoco]'], 'cma': ['cma']},