Merge pull request #554 from WeikeYe-TRI/atomate

Atomate
TRI-AMDD · Jul 5, 2022 · aa31cd0 · aa31cd0
2 parents 2931ff9 + e57f856
commit aa31cd0
Show file tree

Hide file tree

Showing 20 changed files with 1,140 additions and 326 deletions.
diff --git a/.github/workflows/camd-test-main.yml b/.github/workflows/camd-test-main.yml
@@ -27,7 +27,7 @@ jobs:
       run: |
         python -m pip install --upgrade pip
         pip install -r requirements.txt
-        pip install -e .[tests,m3gnet]
+        pip install -e .[tests,m3gnet,atomate]
     - name: pytest
       env:
         MPLBACKEND: "Agg"

diff --git a/Dockerfile b/Dockerfile
@@ -22,4 +22,4 @@ RUN source /opt/conda/bin/activate camd && \
     pip install -r requirements.txt
 
 COPY camd /home/camd/camd
-RUN pip install -e .[proto_dft,m3gnet]
+RUN pip install -e .[proto_dft,m3gnet,atomate]
diff --git a/camd/agent/generic.py b/camd/agent/generic.py
@@ -61,7 +61,7 @@ def get_hypotheses(self, candidate_data, seed_data=None):
 
         self.candidate_data = candidate_data.drop(columns=["target"], axis=1)
         self.seed_data = seed_data
-        X_seed = seed_data.drop(columns=["target"], axis=1, errors='ignore')
+        X_seed = seed_data.drop(columns=["target"], axis=1, errors="ignore")
         y_seed = seed_data["target"]
         steps = [
             ("scaler", StandardScaler()),
@@ -203,8 +203,8 @@ def get_hypotheses(self, candidate_data, seed_data=None):
                         2
                         * np.log(
                             len(self.candidate_data)
-                            * _t ** 2
-                            * np.pi ** 2
+                            * _t**2
+                            * np.pi**2
                             / 6
                             / self.kwargs.get("delta", 0.1)
                         )

diff --git a/camd/agent/m3gnet.py b/camd/agent/m3gnet.py
@@ -42,6 +42,7 @@ def reverse_calcs(retrain_data, data_screening=None):
 
 class M3GNetStabilityAgent(StabilityAgent):
     """Stability agent that uses megnet for estimation of energy"""
+
     def __init__(
         self,
         m3gnet=None,
@@ -147,6 +148,7 @@ def train(self, train_data, epochs=10, kwargs={}):
 
 class M3GNetHypothesisAgent(HypothesisAgent):
     """Generic agent for AL using m3gnet"""
+
     def __init__(self, m3gnet=None, candidate_data=None, seed_data=None, n_query=None):
 
         """

diff --git a/camd/agent/meta.py b/camd/agent/meta.py
@@ -11,31 +11,31 @@
 
 
 REGRESSOR_PARAMS = [
-        {
-            "@class": ["sklearn.linear_model.LinearRegression"],
-            "fit_intercept": [True, False],
-            "normalize": [True, False]
-        },
-        {
-            "@class": ["sklearn.ensemble.RandomForestRegressor"],
-            "n_estimators": [100],
-            "max_features": list(np.arange(0.05, 1.01, 0.05)),
-            "min_samples_split": list(range(2, 21)),
-            "min_samples_leaf": list(range(1, 21)),
-            "bootstrap": [True, False]
-        },
-        {
-            "@class": ["sklearn.neural_network.MLPRegressor"],
-            "hidden_layer_sizes": [
-                # I think there's a better way to support this, but need to think
-                (80, 50),
-                (84, 55),
-                (87, 60),
-            ],
-            "activation": ["identity", "logistic", "tanh", "relu"],
-            "learning_rate": ["constant", "invscaling", "adaptive"]
-        },
-    ]
+    {
+        "@class": ["sklearn.linear_model.LinearRegression"],
+        "fit_intercept": [True, False],
+        "normalize": [True, False],
+    },
+    {
+        "@class": ["sklearn.ensemble.RandomForestRegressor"],
+        "n_estimators": [100],
+        "max_features": list(np.arange(0.05, 1.01, 0.05)),
+        "min_samples_split": list(range(2, 21)),
+        "min_samples_leaf": list(range(1, 21)),
+        "bootstrap": [True, False],
+    },
+    {
+        "@class": ["sklearn.neural_network.MLPRegressor"],
+        "hidden_layer_sizes": [
+            # I think there's a better way to support this, but need to think
+            (80, 50),
+            (84, 55),
+            (87, 60),
+        ],
+        "activation": ["identity", "logistic", "tanh", "relu"],
+        "learning_rate": ["constant", "invscaling", "adaptive"],
+    },
+]
 
 
 AGENT_PARAMS = [
@@ -45,14 +45,14 @@
         "n_members": list(range(2, 5)),
         "hull_distance": list(np.arange(0.05, 0.21, 0.05)),
         "training_fraction": [0.4, 0.5, 0.6],
-        "model": REGRESSOR_PARAMS
+        "model": REGRESSOR_PARAMS,
     },
     {
         "@class": ["camd.agent.agents.AgentStabilityML5"],
         "n_query": [4, 6, 8],
         "hull_distance": [0.05, 0.1, 0.15, 0.2],
         "exploit_fraction": [0.4, 0.5, 0.6],
-        "model": REGRESSOR_PARAMS
+        "model": REGRESSOR_PARAMS,
     },
 ]
 
@@ -73,9 +73,11 @@ def convert_parameter_table_to_dataframe(parameter_table, fillna=np.nan):
 
     """
     df = pd.DataFrame(parameter_table, dtype="int64")
-    df.index = ['-'.join([str(i) for i in row]) for row in np.array(df)]
-    df['agent'] = [parameter_table.hydrate_index(i, construct_object=True)
-                   for i in range(len(parameter_table))]
+    df.index = ["-".join([str(i) for i in row]) for row in np.array(df)]
+    df["agent"] = [
+        parameter_table.hydrate_index(i, construct_object=True)
+        for i in range(len(parameter_table))
+    ]
     df.fillna(fillna)
     return df