From 12c9ed1670261641e53f98f3c8d369490fcd5dff Mon Sep 17 00:00:00 2001
From: Khalim Conn-Kowlessar <kconnkowlessar@gmail.com>
Date: Tue, 22 Jul 2025 19:36:00 +0100
Subject: [PATCH] trying to fix model errors

---
 backend/engine/engine.py  | 2 +-
 backend/engine/handler.py | 1 -
 backend/ml_models/api.py  | 5 +++++
 3 files changed, 6 insertions(+), 2 deletions(-)

diff --git a/backend/engine/engine.py b/backend/engine/engine.py
index 0591eed6..26205328 100644
--- a/backend/engine/engine.py
+++ b/backend/engine/engine.py
@@ -48,7 +48,7 @@ from etl.find_my_epc.RetrieveFindMyEpc import RetrieveFindMyEpc
 logger = setup_logger()
 
 BATCH_SIZE = 5
-SCORING_BATCH_SIZE = 400
+SCORING_BATCH_SIZE = 100
 
 
 def patch_epc(patch, epc_records):
diff --git a/backend/engine/handler.py b/backend/engine/handler.py
index 8fce7f16..fdf48db3 100644
--- a/backend/engine/handler.py
+++ b/backend/engine/handler.py
@@ -11,7 +11,6 @@ def handler(event, context):
     """
     Lambda handler that triggers the model engine for each SQS message.
     """
-    logger.info("Received event: %s", json.dumps(event, indent=2))
     for record in event.get("Records", []):
         try:
             body_dict = json.loads(record["body"])
diff --git a/backend/ml_models/api.py b/backend/ml_models/api.py
index c108f1b7..d814ccf1 100644
--- a/backend/ml_models/api.py
+++ b/backend/ml_models/api.py
@@ -143,6 +143,9 @@ class ModelApi:
                 async with session.post(
                     url, json=payload, headers={"Content-Type": "application/json"}, timeout=120
                 ) as response:
+                    text = await response.text()
+                    if response.status != 200:
+                        logger.error(f"{model_prefix} | Status {response.status} | Body:\n{text}")
                     response.raise_for_status()
                     return await response.json()
             except aiohttp.ClientError as e:
@@ -218,7 +221,9 @@ class ModelApi:
         async with aiohttp.ClientSession() as session:
             for model_prefix in model_prefixes:
                 logger.info(f"Scoring for model prefix: {model_prefix}")
+                logger.info("Uploading scoring data to S3")
                 file_location = self.upload_scoring_data(df, bucket, model_prefix)
+                logger.info("Data uploaded to S3, now making prediction request")
                 # Schedule the prediction request as a coroutine
                 tasks.append(
                     self.predict_async(f"s3://{bucket}/" + file_location, model_prefix)