Create train.py

train.py

@@ -0,0 +1,40 @@
+import pandas as pd
+from fastapi import HTTPException
+import os
+import joblib
+from sklearn.pipeline import Pipeline
+from sklearn.linear_model import LogisticRegression
+from sklearn.model_selection import train_test_split
+from sklearn.feature_extraction.text import TfidfVectorizer
+from sklearn.multioutput import MultiOutputClassifier
+
+from utils import create_text_input
+
+DATA_PATH = "data/synthetic_transactions_samples_5000.csv"
+MODEL_PATH = "models/logreg_model.pkl"
+
+def train_model():
+    try:
+        df = pd.read_csv(DATA_PATH)
+        df = df.fillna("")
+        df["text_input"] = df.apply(create_text_input, axis=1)
+
+        X = df["text_input"]
+        y = df[["Maker_Action", "Escalation_Level", "Risk_Category", "Risk_Drivers", "Investigation_Outcome", "Alert_Status"]]
+
+        X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
+
+        pipeline = Pipeline([
+            ("vectorizer", TfidfVectorizer()),
+            ("classifier", MultiOutputClassifier(LogisticRegression(max_iter=1000)))
+        ])
+        pipeline.fit(X_train, y_train)
+
+        os.makedirs("models", exist_ok=True)
+        joblib.dump(pipeline, MODEL_PATH)
+
+        acc = pipeline.score(X_test, y_test)
+
+        return {"message": "Model trained successfully.", "accuracy": acc}
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))