qwak-sdk 0.5.102__py3-none-any.whl

qwak_sdk/commands/models/init/_logic/template/churn/{{cookiecutter.model_directory}}/main/model.py

@@ -0,0 +1,95 @@
+import os
+
+import numpy as np
+import pandas as pd
+import qwak
+from catboost import CatBoostClassifier, Pool
+from qwak.model.base import QwakModel
+from sklearn.metrics import accuracy_score, f1_score
+from sklearn.model_selection import train_test_split
+
+RUNNING_FILE_ABSOLUTE_PATH = os.path.dirname(os.path.abspath(__file__))
+
+
+class ChurnPrediction(QwakModel):
+    def __init__(self):
+        self.params = {
+            "iterations": int(os.getenv("iterations", 500)),
+            "learning_rate": float(os.getenv("learning_rate", 0.1)),
+            "eval_metric": "Accuracy",
+            "random_seed": int(os.getenv("random_seed", 42)),
+            "logging_level": "Silent",
+            "use_best_model": True,
+        }
+        self.catboost = CatBoostClassifier(**self.params)
+        qwak.log_param(self.params)
+
+    def build(self):
+        df = pd.read_csv(f"{RUNNING_FILE_ABSOLUTE_PATH}/data.csv")
+
+        y = df["churn"]
+        X = df.drop(
+            ["churn", "User_Id", "__index_level_0__", "event date", "Phone"], axis=1
+        )
+
+        categorical_features_indices = np.where(X.dtypes != np.float64)[0]
+        X_train, X_validation, y_train, y_validation = train_test_split(
+            X, y, test_size=0.25, random_state=42
+        )
+
+        train_pool = Pool(X_train, y_train, cat_features=categorical_features_indices)
+        validate_pool = Pool(
+            X_validation, y_validation, cat_features=categorical_features_indices
+        )
+
+        self.catboost.fit(train_pool, eval_set=validate_pool)
+
+        print(
+            "Simple model validation accuracy: {:.4}".format(
+                accuracy_score(y_validation, self.catboost.predict(X_validation))
+            )
+        )
+        print(
+            "Best model validation accuracy: {:.4}".format(
+                accuracy_score(y_validation, self.catboost.predict(X_validation))
+            )
+        )
+
+        y_predicted = self.catboost.predict(X_validation)
+        f1 = f1_score(y_validation, y_predicted)
+
+        qwak.log_metric({"f1_score": f1})
+
+    def schema(self):
+        from qwak.model.schema import ExplicitFeature, InferenceOutput, ModelSchema
+
+        model_schema = ModelSchema(
+            inputs=[
+                ExplicitFeature(name="User_Id", type=str),
+                ExplicitFeature(name="State", type=str),
+                ExplicitFeature(name="Account_Length", type=int),
+                ExplicitFeature(name="Area_Code", type=int),
+                ExplicitFeature(name="Intl_Plan", type=int),
+                ExplicitFeature(name="VMail_Plan", type=int),
+                ExplicitFeature(name="VMail_Message", type=int),
+                ExplicitFeature(name="Day_Mins", type=float),
+                ExplicitFeature(name="Day_Calls", type=int),
+                ExplicitFeature(name="Eve_Mins", type=float),
+                ExplicitFeature(name="Eve_Calls", type=int),
+                ExplicitFeature(name="Night_Mins", type=float),
+                ExplicitFeature(name="Night_Calls", type=int),
+                ExplicitFeature(name="Intl_Mins", type=float),
+                ExplicitFeature(name="Intl_Calls", type=int),
+                ExplicitFeature(name="CustServ_Calls", type=int),
+                ExplicitFeature(name="Agitation_Level", type=int),
+            ],
+            outputs=[InferenceOutput(name="Churn_Probability", type=float)],
+        )
+        return model_schema
+
+    @qwak.api()
+    def predict(self, df):
+        df = df.drop(["User_Id"], axis=1)
+        return pd.DataFrame(
+            self.catboost.predict_proba(df[self.catboost.feature_names_])[:, 1], columns=["Churn_Probability"]
+        )

qwak_sdk/commands/models/init/_logic/template/churn/{{cookiecutter.model_directory}}/tests/it/test_churn.py

@@ -0,0 +1,32 @@
+import pandas as pd
+from qwak.testing.fixtures import real_time_client
+from qwak_inference.realtime_client.client import InferenceOutputFormat
+
+
+def test_realtime_churn(real_time_client):
+    feature_vector = [
+        {
+            "User_Id": 166056434,
+            "State": "AZ",
+            "Account_Length": 140,
+            "Area_Code": 408,
+            "Intl_Plan": 0,
+            "VMail_Plan": 0,
+            "VMail_Message": 0,
+            "Day_Mins": 149.8,
+            "Day_Calls": 134,
+            "Eve_Mins": 164.4,
+            "Eve_Calls": 98,
+            "Night_Mins": 294.7,
+            "Night_Calls": 124,
+            "Intl_Mins": 8.1,
+            "Intl_Calls": 2,
+            "CustServ_Calls": 100,
+            "Agitation_Level": 80,
+        }
+    ]
+
+    survived_probability: pd.DataFrame = real_time_client.predict(feature_vector,
+                                                                  output_format=InferenceOutputFormat.PANDAS)
+
+    assert survived_probability["Churn_Probability"].values[0] > 0.5

qwak_sdk/commands/models/init/_logic/template/credit_risk/cookiecutter.json

@@ -0,0 +1,3 @@
+{
+    "model_directory": "credit_risk"
+}

qwak_sdk/commands/models/init/_logic/template/credit_risk/{{cookiecutter.model_directory}}/main/__init__.py

@@ -0,0 +1,5 @@
+from .model import CreditRisk
+
+
+def load_model():
+    return CreditRisk()

qwak_sdk/commands/models/init/_logic/template/credit_risk/{{cookiecutter.model_directory}}/main/conda.yml

@@ -0,0 +1,11 @@
+name: CreditRisk
+channels:
+  - defaults
+  - conda-forge
+dependencies:
+  - python=3.9
+  - pip=22.2.2
+  - pandas=1.1.5
+  - scikit-learn=0.24.1
+  - catboost=0.26.1
+  - numpy=1.24.0