Review comments

awslabs · Nov 17, 2023 · c06b4c2 · c06b4c2
1 parent f199677
commit c06b4c2
Show file tree

Hide file tree

Showing 2 changed files with 3 additions and 12 deletions.
diff --git a/python/graphstorm/gconstruct/file_io.py b/python/graphstorm/gconstruct/file_io.py
@@ -200,16 +200,15 @@ def read_data_parquet(data_file, data_fields=None):
     """
     table = pq.read_table(data_file)
     data = {}
-    df_table: pd.DataFrame = table.to_pandas()
+    df_table = table.to_pandas()
     assert df_table.shape[0] > 0, \
         f"{data_file} has an empty data. The data frame shape is {df_table.shape}"
 
     if data_fields is None:
         data_fields = list(df_table.keys())
     for key in data_fields:
         assert key in df_table, f"The data field {key} does not exist in {data_file}."
-        val = df_table[key].to_numpy()
-        d = np.array(val)
+        d = df_table[key].to_numpy()
 
         # For multi-dimension arrays, we split them by rows and
         # save them as objects in parquet. We need to merge them

diff --git a/tests/end2end-tests/graphstorm-ec/mgpu_test.sh b/tests/end2end-tests/graphstorm-ec/mgpu_test.sh
@@ -89,15 +89,7 @@ then
 fi
 
 echo "**************dataset: Generated multilabel MovieLens EC, do inference on saved model"
-python3 -m graphstorm.run.gs_edge_classification --inference --workspace $GS_HOME/inference_scripts/ep_infer \
-    --num-trainers $NUM_INFO_TRAINERS --num-servers 1 --num-samplers 0 \
-    --part-config /data/movielen_100k_multi_label_ec/movie-lens-100k.json \
-    --ip-config ip_list.txt --ssh-port 2222 --cf ml_ec_infer.yaml  \
-    --multilabel true --num-classes 6 --node-feat-name movie:title user:feat \
-    --use-mini-batch-infer false --save-embed-path /data/gsgnn_ec/infer-emb/ \
-    --restore-model-path /data/gsgnn_ec/epoch-$best_epoch/ \
-    --save-prediction-path /data/gsgnn_ec/prediction/ --logging-file /tmp/log.txt  \
-    --logging-level debug --preserve-input True
+python3 -m graphstorm.run.gs_edge_classification --inference --workspace $GS_HOME/inference_scripts/ep_infer --num-trainers $NUM_INFO_TRAINERS --num-servers 1 --num-samplers 0 --part-config /data/movielen_100k_multi_label_ec/movie-lens-100k.json --ip-config ip_list.txt --ssh-port 2222 --cf ml_ec_infer.yaml  --multilabel true --num-classes 6 --node-feat-name movie:title user:feat --use-mini-batch-infer false --save-embed-path /data/gsgnn_ec/infer-emb/ --restore-model-path /data/gsgnn_ec/epoch-$best_epoch/ --save-prediction-path /data/gsgnn_ec/prediction/ --logging-file /tmp/log.txt  --logging-level debug --preserve-input True
 
 error_and_exit $?