Start extracting event features.

Calculate threat and challenge means.

features/esm_SAM.py

@@ -0,0 +1,80 @@
+import pandas as pd
+
+import features.esm
+
+QUESTIONNAIRE_ID_SAM = {
+    "event_stress": 87,
+    "event_threat": 88,
+    "event_challenge": 89,
+    "event_time": 90,
+    "event_duration": 91,
+    "event_work_related": 92,
+    "period_stress": 93,
+}
+QUESTIONNAIRE_ID_SAM_LOW = min(QUESTIONNAIRE_ID_SAM.values())
+QUESTIONNAIRE_ID_SAM_HIGH = max(QUESTIONNAIRE_ID_SAM.values())
+
+GROUP_QUESTIONNAIRES_BY = [
+    "participant_id",
+    "device_id",
+    "esm_session",
+    "questionnaire_id",
+]
+# Each questionnaire occurs only once within each esm_session on the same device within the same participant.
+
+
+def extract_stressful_events(df_esm: pd.DataFrame) -> pd.DataFrame:
+    # 0. Select only questions from Stress Appraisal Measure.
+    df_esm_preprocessed = features.esm.preprocess_esm(df_esm)
+    df_esm_sam = df_esm_preprocessed[
+        (df_esm_preprocessed["questionnaire_id"] >= QUESTIONNAIRE_ID_SAM_LOW)
+        & (df_esm_preprocessed["questionnaire_id"] <= QUESTIONNAIRE_ID_SAM_HIGH)
+    ]
+
+    df_esm_sam_clean = features.esm.clean_up_esm(df_esm_sam)
+    # 1.
+    df_esm_event_threat_challenge_mean_wide = calculate_threat_challenge_means(
+        df_esm_sam_clean
+    )
+
+    return df_esm_event_threat_challenge_mean_wide
+
+
+def calculate_threat_challenge_means(df_esm_sam_clean: pd.DataFrame) -> pd.DataFrame:
+    # Select only threat and challenge assessments for events
+    df_esm_event_threat_challenge = df_esm_sam_clean[
+        (
+            df_esm_sam_clean["questionnaire_id"]
+            == QUESTIONNAIRE_ID_SAM.get("event_threat")
+        )
+        | (
+            df_esm_sam_clean["questionnaire_id"]
+            == QUESTIONNAIRE_ID_SAM.get("event_challenge")
+        )
+    ]
+    # Calculate mean of threat and challenge subscales for each ESM session.
+    df_esm_event_threat_challenge_mean = (
+        df_esm_event_threat_challenge.groupby(GROUP_QUESTIONNAIRES_BY)
+        .esm_user_answer_numeric.agg("mean")
+        .reset_index()
+        .rename(columns={"esm_user_answer_numeric": "esm_numeric_mean"})
+    )
+    # Rename questionnaire ID to indicate their names.
+    df_esm_event_threat_challenge_mean[
+        "event_subscale"
+    ] = df_esm_event_threat_challenge_mean.questionnaire_id.astype(
+        "category"
+    ).cat.rename_categories(
+        {
+            QUESTIONNAIRE_ID_SAM.get("event_threat"): "threat_mean",
+            QUESTIONNAIRE_ID_SAM.get("event_challenge"): "challenge_mean",
+        }
+    )
+    # Pivot a table so that each ESM session is represented by one row with threat and challenge means as two columns.
+    df_esm_event_threat_challenge_mean_wide = pd.pivot(
+        df_esm_event_threat_challenge_mean,
+        index=GROUP_QUESTIONNAIRES_BY,
+        columns=["event_subscale"],
+        values=["esm_numeric_mean"],
+    )
+    return df_esm_event_threat_challenge_mean_wide