diff --git a/notebooks/PySpark_linreg_P4.ipynb b/notebooks/PySpark_linreg_P4.ipynb index 4475539..a254ce5 100644 --- a/notebooks/PySpark_linreg_P4.ipynb +++ b/notebooks/PySpark_linreg_P4.ipynb @@ -967,6 +967,7 @@ }, "outputs": [], "source": [ + "# split train test\n", "train_df, test_df = v_df.randomSplit([0.7, 0.3])" ] }, diff --git a/src/PySpark_linreg_P4.py b/src/PySpark_linreg_P4.py index 549b5c1..935631e 100644 --- a/src/PySpark_linreg_P4.py +++ b/src/PySpark_linreg_P4.py @@ -124,7 +124,7 @@ # Log results to S3 conn = boto.s3.connect_to_region("eu-north-1") -bucket = conn.get_bucket("carl-p8") -key = boto.s3.key.Key(bucket, "predictions.txt") +bucket = conn.get_bucket("carl-p8-v2") +key = boto.s3.key.Key(bucket, "/p4_cloud/predictions.txt") key.set_contents_from_string(json.dumps(lr_predictions, indent=2))