From d1e69415cc1239cbcb178cc2cadf1f4f378a2c9a Mon Sep 17 00:00:00 2001
From: anzhengqi <anzhengqi1@huawei.com>
Date: Mon, 26 Sep 2022 21:57:51 +0800
Subject: [PATCH] modify wide_and_deep_multitable network scipts
---
.../wide_and_deep_multitable/src/datasets.py | 21 +++++++++----------
1 file changed, 10 insertions(+), 11 deletions(-)
diff --git a/official/recommend/wide_and_deep_multitable/src/datasets.py b/official/recommend/wide_and_deep_multitable/src/datasets.py
index 1aec7cae9..bb40c7003 100644
--- a/official/recommend/wide_and_deep_multitable/src/datasets.py
+++ b/official/recommend/wide_and_deep_multitable/src/datasets.py
@@ -164,7 +164,6 @@ def _get_h5_dataset(data_dir, train_mode=True, epochs=1, batch_size=1000):
def _get_tf_dataset(data_dir,
- schema_dict,
input_shape_dict,
train_mode=True,
epochs=1,
@@ -186,10 +185,16 @@ def _get_tf_dataset(data_dir,
float_key_list = ["label", "continue_val"]
- columns_list = []
- for key, attr_dict in schema_dict.items():
- print("key: {}; shape: {}".format(key, attr_dict["tf_shape"]))
- columns_list.append(key)
+ columns_list = ["label", "continue_val", "indicator_id", "emb_128_id",
+ "emb_64_single_id", "multi_doc_event_category_id",
+ "multi_doc_event_category_id_mask", "multi_doc_ad_entity_id",
+ "multi_doc_ad_entity_id_mask", "multi_doc_event_entity_id",
+ "multi_doc_event_entity_id_mask", "multi_doc_ad_topic_id",
+ "multi_doc_ad_topic_id_mask", "multi_doc_ad_category_id",
+ "multi_doc_ad_category_id_mask", "multi_doc_event_topic_id",
+ "multi_doc_event_topic_id_mask", "ad_id", "display_ad_and_is_leak",
+ "display_id", "is_leak"]
+ for key in columns_list:
if key in set(float_key_list):
ms_dtype = mstype.float32
else:
@@ -223,7 +228,6 @@ def _get_tf_dataset(data_dir,
print("input_shape_dict start logging")
print(input_shape_dict)
print("input_shape_dict end logging")
- print(schema_dict)
def mixup(a, b, c, d, e, f, g, h, i, j, k, l, m, n, o, p, q, r, s, t, u):
a = np.asarray(a.reshape(batch_size,))
@@ -322,16 +326,11 @@ def create_dataset(data_dir,
create_dataset
"""
if is_tf_dataset:
- with open(os.path.join(data_dir, 'dataformat', "schema_dict.pkl"),
- "rb") as file_in:
- print(os.path.join(data_dir, 'dataformat', "schema_dict.pkl"))
- schema_dict = pickle.load(file_in)
with open(
os.path.join(data_dir, 'dataformat', "input_shape_dict.pkl"),
"rb") as file_in:
input_shape_dict = pickle.load(file_in)
return _get_tf_dataset(data_dir,
- schema_dict,
input_shape_dict,
train_mode,
epochs,
--
GitLab